Significant figures are honesty notation: they say how much of a number you actually know. Writing 2.50 claims more measured precision than 2.5 — and this calculator both counts the sig figs you typed and rounds to however many you need.
The counting rules, compressed
Nonzero digits always count. Zeros between them count (105 has three). Leading zeros never count — they only place the decimal point, so 0.00420 has three sig figs. Trailing zeros count after a decimal point (2.50 → three) but are ambiguous in bare integers like 4,500 — the reason science prefers 4.5 × 10³, and why the calculator flags that case.
Rounding to sig figs vs decimal places
Two different trims. 12,345.6 to two decimal places keeps everything: 12,345.60. To two sig figs it becomes 12,000 — magnitude preserved, precision honestly reduced. 0.0126745 to three sig figs is 0.0127. The mode toggle switches between the trims; the scientific-notation badge shows the unambiguous form either way.
The arithmetic rules everyone mixes up
Multiplication and division: keep as many sig figs as the least precise factor (2.5 × 3.42 = 8.6, not 8.55). Addition and subtraction: keep the fewest decimal places instead (12.11 + 0.3 = 12.4). Different operations, different rules — most lost lab points live right here.
Rounding ties
This tool rounds half up (2.5 → 3), the classroom standard. Statistical work sometimes prefers banker's rounding (half to even) to cancel bias across thousands of roundings — worth knowing when your spreadsheet disagrees with your homework. For spread and precision in datasets, continue to the standard deviation calculator.
Round-half-up convention; inputs parsed to double precision.