Significant Figures Calculator
Enter a number as you would write it. The calculator reports how many significant figures it carries and why, then rounds it to the count you ask for. How you write the number matters: 100, 100. and 1.00 × 10² are three different claims about precision.
Precision is a claim, and it can be overstated by accident
A number written down carries two pieces of information: how large the quantity is, and how well it is known. The first is obvious and the second is invisible unless you know where to look. Significant figures are the notation for the second — the convention that lets 2.5 and 2.50 be different statements about the same measurement, one claiming to be good to a tenth and the other to a hundredth.
This calculator does two things with that convention. It counts the significant figures a number carries as you wrote it, explaining which digits counted and why, and it rounds the number to any count you ask for. Both halves depend on the exact characters typed, which is why the input is a text box rather than a number field: 100 and 100. are different inputs and JavaScript would quietly turn both into the same value.
Every measuring instrument stops somewhere. A ruler marked in millimetres cannot tell you a length to the micron, and writing the answer as 12.4738 mm asserts something the ruler never said. The reverse error is just as common: recording 12 mm when the instrument resolved 12.0 throws away a digit that was genuinely measured.
Calculators make the first error effortless. Divide 1 by 3 and the screen offers ten digits, none of which came from the world. The rule that keeps the arithmetic honest is to carry the digits through the calculation and round at the end, to the precision the weakest input justified — and that requires knowing how many figures each input actually carried.
This tool answers the second question rather than the first. It tells you how many significant figures a value has and rounds it on request. What it deliberately does not do is apply the combination rules to a calculation, because those belong to the operation rather than to any single number.
How to use this calculator
- Type the number exactly as it is written Including a trailing decimal point if there is one, and using e-notation for scientific form. 100, 100. and 1.00e2 are three different inputs and they give three different counts. Copying from a report rather than retyping avoids losing the character that carries the meaning.
- Read the count and the reason beneath it The result states how many figures the number carries and which rule produced that answer — leading zeros discarded, interior digits kept, trailing zeros kept or treated as ambiguous. The reason matters more than the count when the number is one you are about to publish.
- Set how many figures to round to A whole number from 1 to 15. Fifteen is the practical ceiling for double-precision arithmetic; asking for more would print digits that are an artefact of binary storage rather than of the value.
- Check whether scientific notation is suggested When the rounded value needs placeholder zeros — 1230 for three figures, say — the tool offers the scientific form as well, because the plain form cannot distinguish a measured zero from a placeholder.
The counting rules, in the order they apply
First, scientific notation settles the question outright. In 1.00 × 10² only the mantissa carries significance, and every digit written in it counts. That is what the notation is for: it separates the size of the number, which lives in the exponent, from the precision, which lives in the mantissa.
Second, a decimal point makes every digit after the first non-zero one significant. In 0.004520 the leading zeros are placeholders that locate the decimal point, and the digits 4, 5, 2 and the final 0 are all measured — four significant figures. Nobody writes that last zero by accident.
Third, in a whole number with no decimal point, trailing zeros are ambiguous. The 100 in a report might be a count of exactly one hundred, a measurement good to the nearest ten, or a rough figure good to the nearest hundred. This calculator reads it conservatively as one significant figure and says so, because the alternative is to invent precision that was never claimed.
Rounding keeps that many digits from the first non-zero one and rounds the last on the next. It uses the language's own precision formatting rather than a multiply-round-divide, which accumulates error.
What each input means
- x The number, as written — form field “Number”
- Any decimal or scientific-notation value. The exact characters matter: a trailing decimal point, a trailing zero after the point, and the e-notation form all change the count.
- n Figures to round to — form field “Round to how many significant figures”
- A whole number from 1 to 15. Asking for more figures than the value carries does not create precision; it pads with digits the measurement never supported.
Worked examples
Every number below is produced by the same calculation engine the tool above runs. Nothing here is typed by hand, so the walkthrough cannot drift from what you get when you enter the same values yourself.
A laboratory measurement to two figures
A concentration is recorded as 0.004520 grams per litre and a summary table has room for two significant figures. The question is what to print, and whether anything is lost by printing it.
Inputs Number = 0.004520, Round to how many significant figures = 2
- As written 0.004520 has 4 significant figure(s)
- Why A decimal point is present, so every digit after the first non-zero one counts — including trailing zeros.
- Rounding Keep 2 digit(s) starting from the first non-zero one, then round the last kept digit on the next one.
- Result 0.004520 → 0.0045
- Write it in scientific notation 4.5e-3 — the plain form would need placeholder zeros, and those are exactly the ambiguous ones.
Result 4 significant figure(s); rounded to 2: 0.0045
The count comes back as four. The two zeros before the 4 are placeholders and contribute nothing; the 4, the 5, the 2 and the final 0 are all digits the instrument resolved.
Rounded to two figures the value becomes 0.0045. That discards the 2 and the trailing 0, which is a real loss of information — the table is now claiming a tenfold coarser measurement than was taken. Whether that matters depends on what the table is for, but the tool makes the loss visible rather than silent.
The ambiguous hundred
A field note reads 100 metres. Somebody downstream needs to know whether that is a pace count, a tape measurement, or a rounded estimate — and the number alone cannot say.
Inputs Number = 100, Round to how many significant figures = 2
- As written 100 has 1 significant figure(s)
- Why Trailing zeros in a whole number with no decimal point are ambiguous. They are counted as NOT significant here, which is the conservative reading; write 100. or 1.00 × 10² to mean three.
- Rounding Keep 2 digit(s) starting from the first non-zero one, then round the last kept digit on the next one.
- Result 100 → 1.0e+2
- Write it in scientific notation 1.0e+2 — the plain form would need placeholder zeros, and those are exactly the ambiguous ones.
Result 1 significant figure(s); rounded to 2: 1.0e+2
The count is reported as one significant figure, with the ambiguity named explicitly. That is the conservative reading: it assumes the least the writer might have meant.
The remedy is notation rather than arithmetic. Writing 100. commits to three figures and 1.0 × 10² commits to two. Both are available and both are unambiguous; the bare 100 is the only form that cannot be pinned down, which is why it keeps causing this problem.
Reading the result
The count, and what it is a count of
The figure returned is a property of the written number, not of the quantity. The same length can be written 1.2 m, 120 cm and 1200 mm and get counts of two, two and two — or two, three and four, if the writer intended those trailing zeros. The count measures the claim, and the claim lives in the notation.
When you would use this
Reporting a measurement
Instrument readouts often show more digits than the instrument resolves. Rounding to the manufacturer's stated resolution, and knowing how many figures that is, is the difference between a reported value and an implied claim about the equipment.
Assumptions and limitations
What this calculator assumes
- The number is a measurement or a derived value, not an exact count. A count of 12 eggs has infinite significant figures because it is not a measurement at all, and no rule here applies to it.
- The written form reflects what was intended. The tool reads notation literally, which is the only thing it can do; if a trailing zero was typed carelessly the count will be wrong in the direction of the typing.
Where it stops being the right tool
- It does not apply the combination rules. A product takes the fewest significant figures of its inputs and a sum takes the fewest decimal places, and both are properties of the calculation rather than of a value — this tool works on one number at a time.
Common mistakes
Counting the leading zeros in 0.0045
Why it happens. They are visually prominent, and in a number like 1.0045 zeros in that position genuinely do count.
How to avoid it. Start at the first non-zero digit. Everything left of it places the decimal point, which is why 0.0045 and 4.5 × 10⁻³ both carry two figures.
Sources and further reading
- The International System of Units (SI Brochure), 9th edition Bureau International des Poids et Mesures Cited for the convention on expressing the precision of a measured quantity and the use of scientific notation to avoid ambiguous trailing zeros.
Frequently asked questions
Are trailing zeros after a decimal point always significant?
Yes. A zero written after the decimal point serves no purpose except to record precision, so 2.50 claims a measurement good to a hundredth while 2.5 claims only a tenth.
Is zero itself significant?
Zero as a value has no significant figures to count — there are no non-zero digits to start from. Zeros inside a number are a different matter and are counted normally.
Why does this tool read 100 as one significant figure rather than three?
Because one is the least the writer might have meant, and assuming more would manufacture precision. The output says the reading is conservative and names the two notations that remove the doubt.