Two valid measures describe different movement
Percentage points measure the direct arithmetic difference between two percentages: new rate − original rate. Relative percent change measures that difference relative to the original rate: (new rate − original rate) ÷ original rate × 100. Percentage points stay in the same scale as the displayed rates, while relative change answers how large the movement is compared with where the rate began.
The distinction matters whenever the inputs are themselves percentages, including interest rates, survey support, conversion rates, market share, unemployment rates, and test pass rates. Saying only that a rate “rose 5%” can be ambiguous. State “5 percentage points” for direct movement or “25% relative increase” for a baseline-scaled comparison.
The essential 20% to 25% example
Input: an original rate of 20% and a new rate of 25%. Formula: point change = new rate − original rate; relative change = difference ÷ original rate × 100. Calculation: 25% − 20% = 5 percentage points, while 5 ÷ 20 × 100 = 25%. Result: the rate increased by 5 percentage points and increased by 25% relative to its original level.
Interpretation: both statements are correct. They are not two rounding versions of one answer. Imagine 1,000 otherwise comparable opportunities: 20% corresponds to 200 outcomes and 25% to 250, an increase of 50. That gain is 5 per 100 opportunities, hence 5 percentage points, and 50 relative to the original 200 is 25%, hence the relative increase.
Worked example: an interest rate
Input: a stated annual interest rate moves from 3.0% to 4.5%. Formula: subtract for points, then divide that difference by 3.0% for relative change. Calculation: 4.5% − 3.0% = 1.5 percentage points; 1.5 ÷ 3.0 × 100 = 50%. Result: the rate rose 1.5 percentage points, which is a 50% relative increase in the rate.
Interpretation: this does not say that a borrower's payment, total interest cost, investment return, or account balance rises by 50%. Those outcomes depend on principal, compounding, term, fees, timing, and product rules. It describes only the quoted rate. In finance, “basis points” may also be used: one percentage point equals 100 basis points, so 1.5 percentage points equals 150 basis points.
Worked example: survey support
Input: support in a comparable survey changes from 40% to 46%. Formula: subtract the rates for points and divide the 6-point gap by the original 40% for relative change. Calculation: 46% − 40% = 6 percentage points; 6 ÷ 40 × 100 = 15%. Result: reported support increased by 6 percentage points, or 15% relative to the earlier share.
Interpretation: neither calculation determines whether the movement is statistically significant. Sample design, sample size, weighting, question wording, nonresponse, field dates, and uncertainty affect that judgment. Percentage-point language is often the clearest description of topline rate movement, while relative change may help explain scale. Identify the population and survey wave rather than implying that all respondents or all residents changed their views.
Worked example: conversion rate performance
Input: a website records 320 conversions from 4,000 visits in one period and 450 from 5,000 visits in another. First compute each rate: 320 ÷ 4,000 × 100 = 8%; 450 ÷ 5,000 × 100 = 9%. Percentage-point change: 9% − 8% = 1 percentage point. Relative change: 1 ÷ 8 × 100 = 12.5%. Result: conversion rate rose 1 percentage point, a 12.5% relative increase.
The conversion count itself rose from 320 to 450, or 40.625%, but traffic also changed. That count change is not the same as the rate change. A useful report gives numerators and denominators alongside the two rate descriptions, checks comparable attribution windows and traffic definitions, and avoids crediting the increase to a campaign without evidence.
Zero baselines and falling rates
Percentage-point change remains defined when the original rate is zero. From 0% to 3%, the movement is +3 percentage points. Relative percent change is undefined because its denominator is the original 0%. Do not report an infinite increase. State that the rate moved from zero to 3%, give the 3-point change, and include underlying counts so readers can judge scale.
For a decline from 25% to 20%, the point change is −5 percentage points. Relative change is (20 − 25) ÷ 25 × 100 = −20%. If describing the magnitude, say a 20% relative decrease. Reversing the direction from 20% to 25% gives a 25% increase, illustrating again that relative change depends on the original baseline even though the point gap has magnitude 5 in both directions.
Rates can exceed 100%, shares usually cannot
Many proportions, such as the share of respondents selecting one exclusive option, lie between 0% and 100%. Some rates are not bounded in the same way: a growth rate or a rate per unit can exceed 100% depending on its definition. Percentage-point subtraction remains arithmetic, but interpretation depends on whether the underlying measure really is a proportion, a probability, or another kind of rate.
Always preserve the rate's definition and time basis. A monthly rate and annual rate cannot be compared by direct subtraction without first establishing a valid conversion. Likewise, two survey percentages from different questions or populations are not made comparable merely because both include percent signs. The labels, denominators, and measurement methods must align.
Calculator use and precise reporting
The Percentage Calculator can apply a known rate to a number and can show the average-based percentage difference between two peer values. It does not directly calculate percentage points, derive a component rate from part and whole, or calculate directional relative change. For 20% to 25%, subtract the displayed rates for the 5-point movement and divide that gap by the original 20% for the 25% relative change. The tool cannot determine whether two rates use comparable populations or definitions.
Avoid writing “increased by 5%” when you mean five percentage points, dividing by 100 twice, using the new rate as the relative-change denominator, or applying the relative change in a rate directly to a payment or population count. A complete statement names both rates, the point movement, the relative movement when useful, and the underlying period or group. That wording lets readers understand the scale without guessing which measure was intended.
Choose the measure that serves the reader
Percentage-point movement is usually the most direct way to compare rates on the same bounded scale. Relative change adds useful context when the starting rate is not zero and readers need to understand proportional growth. A one-point rise from 2% to 3% is a 50% relative increase, while a one-point rise from 50% to 51% is only a 2% relative increase. The identical point movement can therefore have very different relative sizes.
Large relative changes from small baselines deserve the endpoints and underlying counts. A rate moving from 0.1% to 0.2% doubles, but it rises only 0.1 percentage point. Neither description is inherently better; each highlights a different scale. Publishing both can be informative when space permits, provided the wording remains precise and the data populations, observation periods, and rate definitions are genuinely comparable.
Signs also carry meaning. A negative point change indicates that the new rate is lower; a negative relative change reports a decline relative to the original rate. If an audience may confuse a negative sign with a negative rate, write the direction in words. “Fell 5 percentage points, from 25% to 20%” is much harder for a general reader to misread than an isolated value of −5 alone.