Every zero in a number like 47,000,000 or 0.0000392 is technically load-bearing — it fixes where the decimal point sits — but visually those zeros carry no actual information, and a human eye can lose count of them shockingly fast. Scientific notation's whole purpose is separating a number into the part that genuinely matters (its meaningful digits, the "coefficient," squeezed to a value between 1 and 10) and the part that just tracks scale (a power of 10), so nobody has to eyeball a long zero-count ever again.
**Peeling a large number down to its coefficient.** Sliding the decimal point left, one place at a time, until only a single nonzero digit remains in front of it — the number of slides it took becomes the exponent, since each slide is undoing one factor of 10 you're about to restore via the power. For 47,000,000: seven slides left lands on 4.7, so the seven slides get restored as ×10⁷, giving 4.7 × 10⁷.
**Doing the same thing for a small decimal, sliding the other direction.** Sliding right instead of left, since the meaningful digit now sits *after* several placeholder zeros rather than before them — each rightward slide earns a negative exponent instead of a positive one, since you're moving toward, not away from, where the value actually starts. For 0.0000392: five slides right lands on 3.92, so 0.0000392 = 3.92 × 10⁻⁵.
**Reversing the process is simply un-peeling what peeling did.** Since a positive exponent recorded how many places the decimal slid left, restoring the number means sliding it back right by that same count; a negative exponent, having recorded a rightward slide, gets undone by sliding left. Worked example: 6.5 × 10⁵ unwinds by sliding five places right: 650,000. Worked example: 8.1 × 10⁻⁴ unwinds by sliding four places left: 0.00081.
**Why the coefficient must be between 1 and 10.** This is what makes scientific notation a single, unambiguous standard form rather than one of many equivalent-but-different-looking representations. 82,000 could technically be written as 82 × 10³ or 8.2 × 10⁴ — both are mathematically correct, but only 8.2 × 10⁴ follows the "exactly one nonzero digit before the decimal point" rule that defines proper scientific notation.
**Multiplying and dividing in scientific notation.** Multiply the coefficients and add the exponents: (3 × 10⁴) × (2 × 10³) = (3×2) × 10^(4+3) = 6 × 10⁷. Divide the coefficients and subtract the exponents: (8 × 10⁶) ÷ (2 × 10²) = (8÷2) × 10^(6−2) = 4 × 10⁴. Both operations rely directly on the exponent rules covered in Understanding Exponents and Powers.
**Adding and subtracting in scientific notation requires an extra step.** The exponents must match before you can combine coefficients directly. To add (5 × 10⁴) + (3 × 10³), first rewrite one term so both share the same exponent: 3 × 10³ = 0.3 × 10⁴. Then add coefficients: 5 × 10⁴ + 0.3 × 10⁴ = 5.3 × 10⁴. Skipping this exponent-matching step is the most common error when combining scientific-notation values.
**Where this actually gets used, and why the habit stuck.** Physics, chemistry, and astronomy trade constantly in numbers that are absurd to write out longhand — atomic masses, distances between galaxies, the charge on a single electron — and scientific notation became the field's default not out of tradition but because the alternative genuinely doesn't scale; a journal article rewriting every constant with its full string of zeros would be both harder to typeset and far easier to misread by a stray digit.
**A cousin convention worth knowing about: engineering notation.** Rather than insisting on exactly one digit before the decimal point, engineering notation locks the exponent to a multiple of 3, deliberately lining up with metric prefixes — kilo-, mega-, milli-, micro- — so an engineer reading "82 × 10³" can map it straight onto "82 kilo-something" without doing any mental conversion. It's a genuinely different formatting rule from strict scientific notation, not just a stylistic variant, so it's worth noticing which one a given technical document has actually adopted.
**Common mistakes.** A coefficient left outside the 1-to-10 range — writing 47 × 10⁶ rather than properly normalizing it to 4.7 × 10⁷ — is technically off-standard, even though the value itself is correct. Sign errors on the exponent are the other frequent slip: attaching a positive exponent to a number under 1, or a negative one to something 10 or larger, gets the direction of the shift backwards.
**Worked example: convert Earth's approximate mass, 5,970,000,000,000,000,000,000,000 kg, to scientific notation.** Counting digits to move the decimal point left until one nonzero digit remains before it: 24 places. Result: 5.97 × 10²⁴ kg — a case where scientific notation's compactness advantage over standard notation is dramatic; writing out 24 zeros by hand invites miscounting in a way 5.97 × 10²⁴ simply doesn't.
**Worked example: multiply two very different-sized scientific-notation numbers.** (6 × 10⁻⁹) × (4 × 10¹²). Multiply coefficients: 6×4=24. Add exponents: −9+12=3. Result: 24 × 10³ — but this isn't proper scientific notation yet, since the coefficient (24) isn't between 1 and 10. Adjust: 24 × 10³ = 2.4 × 10¹ × 10³ = 2.4 × 10⁴. This adjustment step — renormalizing the coefficient back into the 1-10 range after a multiplication — is a commonly missed final step.
**Comparing the size of two numbers in scientific notation, a fast use case.** Between 3.2 × 10⁸ and 9.1 × 10⁷, the first has the larger exponent (8 versus 7), so it's larger regardless of the coefficients — 3.2 × 10⁸ = 320,000,000, clearly bigger than 9.1 × 10⁷ = 91,000,000. Comparing exponents first, and only comparing coefficients if the exponents match, is a fast way to rank several scientific-notation values without converting any of them back to standard form.
**Worked example: divide two scientific-notation numbers with a negative-exponent result.** (2 × 10³) ÷ (5 × 10⁷). Divide coefficients: 2÷5=0.4. Subtract exponents: 3−7=−4. Result: 0.4 × 10⁻⁴ — not yet proper form, since the coefficient must be at least 1. Adjust: 0.4 × 10⁻⁴ = 4 × 10⁻¹ × 10⁻⁴ = 4 × 10⁻⁵. As with the earlier multiplication example, renormalizing the coefficient back into the 1-10 range is an easy final step to forget.
**Recognizing scientific notation in calculator and spreadsheet displays.** Many calculators and spreadsheet programs automatically switch to scientific notation for very large or very small results, often displayed with an "E" or "e" instead of "×10" — a value shown as 4.7E+08 means 4.7 × 10⁸, and 3.2E−05 means 3.2 × 10⁻⁵. Recognizing this shorthand is worth knowing since it's the form scientific notation actually appears in most often in everyday digital tools, not the "×10" notation used in textbooks.
For how significant figures and precision interact with scientific notation specifically, see How to Round Numbers Correctly.