To calculate an unknown concentration from a standard curve, you measure your unknown sample’s signal (such as absorbance), then use the equation of your curve’s best-fit line to solve for concentration. For a linear curve following the format y = mx + b, you plug your signal in for y and solve for x: concentration = (y − b) / m. The whole process takes just a few minutes once your curve is built, but getting reliable results depends on how well you set up your standards and how carefully you handle the math.
Why Signal Relates to Concentration
Standard curves work because of a predictable physical relationship: as the amount of a substance in solution increases, the signal it produces increases proportionally. In spectrophotometry, this relationship is described by Beer’s Law, which states that absorbance equals the product of a constant (related to the substance and the path length of light) multiplied by concentration. Beer demonstrated in 1852 that when you keep the measurement conditions constant, doubling the concentration doubles the absorbance. That proportional relationship is what makes a straight line on your graph and what lets you work backward from signal to concentration.
This linear relationship holds across a useful range but not forever. At very high or very low concentrations, the curve bends. Your standard curve defines the window where the math is trustworthy.
Building the Standard Curve
You start by preparing a series of solutions with known concentrations of your analyte. These are your standards. A typical set includes five to eight concentrations spanning the range you expect your unknowns to fall within, plus a blank (zero concentration) to establish your baseline signal. Space the concentrations evenly or, if you need to cover several orders of magnitude, use a serial dilution series and plan to plot on a logarithmic scale.
Before preparing your dilutions, make a rough estimate of where your unknown concentration likely falls. If you think it could range from, say, 0.7 to 7 mg/mL, your standards need to bracket that entire range. It’s better to overshoot slightly on both ends than to discover your unknown falls outside the curve. You then measure each standard under the same conditions you’ll use for your unknowns, recording the signal (absorbance, fluorescence, optical density, or whatever your assay produces) for every known concentration.
Plot concentration on the x-axis and signal on the y-axis. Add a best-fit line, either by hand or using software, and note the equation of that line.
The Core Calculation
For a linear standard curve, your best-fit line gives you an equation in the form y = mx + b, where y is the signal, m is the slope, x is the concentration, and b is the y-intercept. To find the concentration of your unknown, rearrange the equation:
x = (y − b) / m
Here’s a concrete example. Suppose your best-fit line equation is y = 0.042x + 0.005, and your unknown sample gives an absorbance reading of 0.350. Plug in:
x = (0.350 − 0.005) / 0.042 = 0.345 / 0.042 = 8.21
Your unknown concentration is 8.21 in whatever units your standards used (mg/mL, µg/mL, etc.).
Accounting for Dilutions
If you diluted your unknown sample before measuring it, the concentration you just calculated is the diluted concentration. To find the original concentration, multiply by the dilution factor. If you made a 1:5 dilution (one part sample, four parts diluent), your dilution factor is 5. So the original concentration would be 8.21 × 5 = 41.05 in your working units.
This step is easy to forget, especially when you’ve done multiple dilutions, but skipping it is one of the most common sources of error in lab calculations.
Checking Your Curve’s Quality
Before trusting your result, check how well the line fits your data. The R² value (coefficient of determination) tells you how much of the variation in your signal is explained by concentration. An R² of 1.0 means perfect fit. For most analytical assays, you want R² ≥ 0.95. In practice, a well-prepared standard curve in spectrophotometry often hits 0.99 or higher. If your R² is below 0.95, something went wrong with your standards, your pipetting, or your instrument, and the concentrations you calculate from that curve won’t be reliable.
Also confirm that your unknown’s signal falls within the range of your standards. Extrapolating beyond the highest or lowest standard is unreliable because you have no evidence the linear relationship holds outside that range. If your unknown’s reading is too high, dilute it and re-measure. If it’s too low, you may need to concentrate your sample or use a more sensitive assay.
When the Curve Isn’t a Straight Line
Not all standard curves are linear. Immunoassays like ELISAs typically produce S-shaped (sigmoidal) curves, where the signal plateaus at both low and high concentrations. A straight-line equation won’t fit this data well, so these assays use a four-parameter logistic (4PL) regression instead. The 4PL model accounts for the upper plateau, lower plateau, the midpoint, and the steepness of the curve. If the S-shape is asymmetrical (one end bends differently than the other), a five-parameter logistic (5PL) model adds a correction for that asymmetry.
You won’t solve a 4PL equation by hand. Software built into plate readers, or standalone programs, fits the curve and calculates unknown concentrations automatically. The principle is the same as the linear method: the software finds the concentration on the x-axis that corresponds to your measured signal on the y-axis. You just can’t do the algebra on paper.
Doing It in Excel
For a linear standard curve, Excel makes the calculation straightforward. Enter your known concentrations in one column and the corresponding signal values in another. Select both columns, insert a scatter chart, then right-click the data points and choose “Add Trendline.” Select “Linear” and check the boxes to display the equation and R² value on the chart.
Once you have the equation, you can calculate unknown concentrations manually using the rearranged formula. Alternatively, Excel’s TREND function can automate this. Set up a cell with your unknown’s signal value, and TREND will interpolate the corresponding concentration based on your standard data. For a single unknown, the FORECAST function works similarly and may be more intuitive: you give it the signal value and point it to your known data, and it returns the predicted concentration.
Common Pitfalls That Skew Results
One significant issue in sandwich immunoassays is the hook effect. At very high analyte concentrations, the signal paradoxically decreases instead of continuing to rise. This happens because excess analyte saturates the capture antibodies, blocking the labeled detection molecules from binding properly. The result is a falsely low reading that can look like a normal, mid-range concentration. If you suspect your sample might have a very high concentration, run it at multiple dilutions. If a more diluted sample gives a higher calculated concentration than a less diluted one, the hook effect is likely at play.
Other common problems include using expired or improperly stored standards, inconsistent pipetting volumes across the dilution series, and air bubbles in the measurement path. Each of these introduces scatter in your data points, lowers your R² value, and makes your calculated concentrations less accurate. Running each standard in duplicate or triplicate and averaging the signals helps reduce the impact of random pipetting errors.
Detection Limits of Your Curve
Every standard curve has a floor below which it can’t reliably measure. The limit of detection (LoD) is the lowest concentration your assay can distinguish from a blank sample. It’s calculated using the variability of blank measurements and the variability of a low-concentration sample. Specifically, you take the mean signal from blank replicates, add 1.645 times the standard deviation of those blanks to get the limit of blank (LoB), then add 1.645 times the standard deviation of a known low-concentration sample. The 1.645 multiplier corresponds to a 95% confidence threshold.
The limit of quantitation (LoQ) is sometimes higher than the LoD. It’s the lowest concentration where your measurement is not only detectable but also precise enough to be meaningful. If your unknown’s signal falls near or below the lowest standard on your curve, report it as below the limit of quantitation rather than calculating a number that carries little confidence.