Significant Figures Calculator

How to use the significant figures calculator

Type a number and read its significant figures. The input is deliberately a text field rather than a number field, because the distinction this page exists to make is destroyed the moment a computer parses it: 1.400 and 1.4 are the same stored value and different measurements.

Significant figures are a claim about how precisely something was measured. A length of 1.4 metres says the measurement was good to a tenth; 1.400 says it was good to a millimetre. Both describe the same distance, and only one of them justifies quoting a millimetre in the answer.

The four rules are about position and nothing else. A zero before the first non-zero digit is a placeholder holding the decimal point in place, so it carries no information. A zero between two digits was measured, so it does. A zero at the end was measured only if a decimal point is there to say so.

Rounding is the other half of the job, and it has a consequence worth watching. Rounding 1234 to two significant figures gives 1.2e+3 rather than 1200, and that is deliberate: writing 1200 would reintroduce exactly the ambiguity the rounding was meant to resolve.

The two rules that get mixed up

Significant figures govern multiplication and division. Decimal places govern addition and subtraction. They are different rules producing different answers, and applying the wrong one is the commonest error after miscounting zeros.

The reason is worth knowing rather than memorising. In a sum, an uncertainty of a tenth stays an uncertainty of a tenth. In a product, a one percent uncertainty stays one percent, so what matters is the proportion each figure was known to rather than the column it sat in.

What a computer cannot store about your number

Numbers in software are stored as values, not as the text you typed. 1.400 and 1.4 become the same double the moment they are parsed, and every trailing zero is gone. That is why this page reads a string and why any spreadsheet loses the distinction silently.

That last one compounds quietly. Rounding at every step introduces a fresh error each time, and those errors accumulate rather than cancel. Carrying full precision through a calculation and rounding once at the end is both easier and more accurate, and it is what every laboratory convention asks for.

The formula, worked line by line

Four position rules for counting, and one rule each for the two kinds of arithmetic. There is no formula as such — it is a set of conventions about what a written number claims.

Reading them that way makes the ambiguous case obvious rather than annoying: the notation simply does not carry the information, so no rule can recover it.

leading zeros:  never significant     0.0042 -> 2
captive zeros:  always significant    1004   -> 4
trailing, with a decimal point:       1.400  -> 4
trailing, without one:                1400   -> ambiguous
multiplying: answer takes the FEWEST significant figures
adding:      answer takes the FEWEST decimal places

Why the rules differ for adding and multiplying

Because uncertainty behaves differently. In a sum, an uncertainty of a tenth remains a tenth regardless of how large the numbers are, so the limiting factor is the decimal column. In a product, a one percent uncertainty stays one percent, so the limiting factor is the proportion the number was known to.

Where this matters outside a classroom

The habit that matters most is the simplest: carry full precision through the working and round once at the end. Every intermediate rounding is a fresh error, and they add up rather than cancel out.

Questions people ask