A megabyte (MB) is bigger than a kilobyte (KB). In the decimal system used on drive labels, 1 MB equals 1,000 KB. In the binary system used by Windows, 1 MB equals 1,024 KB. Either way, a megabyte is roughly a thousand times larger, so any file measured in MB holds far more data than one measured in KB.
The direction of the answer is fixed, but the exact multiplier is not. That gap between 1,000 and 1,024 is where most storage arguments, billing disputes, and "missing capacity" complaints begin. Understanding why two multipliers exist, and when each applies, matters more than memorizing a single number.
The prefix mechanism: why kilo means two different things
The metric system defines kilo as exactly 1,000. That definition dates to the 1795 adoption of the metric system in France, and it applies to kilograms, kilometers, and kilobytes alike. A decimal kilobyte is therefore 1,000 bytes, and a decimal megabyte is 1,000,000 bytes.
Computers complicate this because they count in powers of two. Memory addresses are binary, so engineers naturally sized storage in chunks of 2^10 (1,024) rather than 10^3 (1,000). Early terminology borrowed the metric prefixes for these binary quantities, so "kilobyte" came to mean 1,024 bytes in casual computing use. The ambiguity was baked in from the start.
The International Electrotechnical Commission (IEC) tried to close the gap in 1998 by introducing distinct binary prefixes. A kibibyte (KiB) is 1,024 bytes, a mebibyte (MiB) is 1,024 KiB, and so on. The IEC standard has been widely adopted in engineering documentation but only partially in consumer software. Windows still displays binary values under decimal labels, which is the single largest source of the confusion.
So the mechanism is simple at its core. One prefix, two counting bases, and decades of inconsistent labeling. The megabyte is always the larger unit; only the size of the jump changes.
Quantified context: how large the gap actually gets
At the kilobyte-to-megabyte step, the discrepancy is 2.4 percent. A file Windows calls 1 MB (1,024 KB) is 2.4 percent larger than the 1,000 KB a decimal label implies. That is small enough to ignore in daily use.
The error compounds as units grow. Each successive binary step multiplies by 1,024 instead of 1,000, adding about 2.4 percent per level. By the gigabyte step, a binary GB is 1,073,741,824 bytes against 1,000,000,000 decimal, a 7.4 percent difference. At the terabyte step the gap reaches roughly 10 percent, and at the petabyte step it approaches 13 percent.
This is why a 1 TB drive shows about 931 GB in Windows. The manufacturer sells 1,000,000,000,000 bytes using decimal definitions, and the operating system divides by 1,024 three times. No space is missing; the two parties are simply counting with different rulers.
Practical anchor points help. A plain-text email runs about 5 to 20 KB. A typical smartphone photo lands between 2 and 5 MB, so one photo equals roughly 2,000 to 5,000 KB. A one-minute 1080p video clip is often 50 to 150 MB. These figures make the thousand-fold relationship concrete.
Comparison table: units, definitions, and typical use
| Unit | Decimal value | Binary value | Common label context |
|---|---|---|---|
| Kilobyte (KB) | 1,000 bytes | 1,024 bytes | Small text files, emails |
| Megabyte (MB) | 1,000,000 bytes | 1,048,576 bytes | Photos, songs, documents |
| Gigabyte (GB) | 1,000,000,000 bytes | 1,073,741,824 bytes | Drive capacity, video |
| Kibibyte (KiB) | 1,024 bytes | 1,024 bytes | IEC-standard binary label |
| Mebibyte (MiB) | 1,048,576 bytes | 1,048,576 bytes | Engineering documentation |
The table makes the core point visible: KB and MB are ordered by size regardless of which column you read. The binary column is always slightly larger than the decimal column at the same label. Only the KiB and MiB rows avoid the discrepancy entirely because they carry unambiguous prefixes.
Edge cases: networking, marketing, and the 1,024 trap
Network speeds use a different convention than storage. Internet throughput is measured in bits per second, not bytes, and the lowercase b versus uppercase B distinction is critical. A connection marketed as 100 Mbps delivers about 12.5 MB per second, because eight bits make one byte. Confusing megabits with megabytes produces an eight-fold error, far larger than the 2.4 percent binary gap.
Storage manufacturers follow decimal definitions for good reason. Drive capacity scales linearly with cost, so selling a "1 TB" drive as 1,000,000,000,000 bytes is a clean, defensible claim. Operating systems that report binary values under the same labels create the illusion of lost space. macOS switched to decimal reporting in 2009 with Snow Leopard, which is why the same drive appears larger on a Mac than on Windows.
The counterintuitive detail: a file labeled "1 MB" in Windows is technically 1 MiB by IEC standards, and a file labeled "1 MB" on a Mac is exactly 1,000,000 bytes. The same label describes different quantities depending on the platform. Professionals who move files between systems learn to check byte counts rather than trust labels, because label-based comparisons can be off by several percent across a large archive.
A professional caveat follows from this. When capacity or billing depends on exact byte counts, such as cloud storage quotas or data-cap billing, always confirm which definition the provider uses. The difference between decimal and binary accounting on a 10 TB plan can exceed 900 GB, which is enough to matter for cost planning.
Misconceptions that persist
The most common misconception is that KB is bigger than MB. It is not, and the ordering follows the prefix: kilo is smaller than mega, just as a kilogram is lighter than a megagram. The confusion arises because people see larger numbers attached to KB labels and assume the unit is larger.
A second misconception treats the 1,024 figure as universal. It applies to binary reporting, not to all contexts. Drive labels, network speeds, and many modern operating systems use decimal math. Assuming 1,024 everywhere produces small errors that grow with file size.
A third misconception holds that "missing" drive space indicates a defect or deceptive marketing. In most cases the capacity is exactly as advertised in decimal terms, and the discrepancy is a reporting convention. Checking byte counts and the operating system's definition resolves the question without a support ticket.
Red flag: a storage provider that will not state whether its quotas use decimal or binary definitions. Red flag: network marketing that mixes Mbps and MB/s in the same sentence without clarifying the difference. Red flag: any comparison of file sizes across Windows and macOS without confirming the reporting base. Red flag: billing disputes resolved by label rather than by exact byte count.
Each warning points to the same underlying risk: a label without a defined counting base is ambiguous, and ambiguity in high-volume contexts compounds into real money or real lost capacity.
Practical actions: resolving the question in your own files
Start by checking the actual byte count. On Windows, right-click a file, choose Properties, and read the "Size" and "Size on disk" fields in bytes. On macOS, use Get Info and look at the byte figure. The byte count is unambiguous and settles any KB-versus-MB comparison immediately.
When comparing capacities, convert both figures to bytes before judging. A quick rule: multiply a decimal MB figure by 1,000,000 and a binary MB figure by 1,048,576. Then compare the results directly. This removes label ambiguity in one step.
Set a threshold for when the gap matters. Below about 100 MB, the decimal-versus-binary difference is under 5 MB and rarely worth worrying about. Above 1 TB, the gap exceeds 90 GB and deserves attention in any capacity plan. For billing or quota decisions, treat the byte count as the source of truth.
Adopt the IEC labels in your own documentation. Writing KiB and MiB when you mean binary values prevents readers from assuming decimal math. This is standard practice in engineering and removes the ambiguity that causes most confusion in the first place.
Verify byte counts before comparing any two files. Convert capacities to bytes before judging "missing" space. Treat the 100 MB and 1 TB marks as thresholds for when the gap matters. Use KiB and MiB labels in technical writing to eliminate ambiguity.
The steps are simple, but they resolve the question definitively. The megabyte is always bigger than the kilobyte; the byte count tells you by exactly how much.
Limitations and what this explanation cannot settle
This explanation covers the stable mechanism of decimal and binary prefixes and the standard reporting conventions. It does not resolve every vendor's labeling choice, because some providers use hybrid or proprietary definitions that are not publicly documented. Where a provider does not publish its counting base, only the byte count on a specific file or plan can settle the question.
The compounding percentages cited here are arithmetic consequences of the 1,024-versus-1,000 ratio, not measured performance figures. They describe label discrepancies, not real differences in storage physics or transfer speed. Real-world throughput depends on hardware, protocol, and workload, which this article does not quantify.
Finally, the practical thresholds are rules of thumb, not standards. A 100 MB cutoff for when the binary gap matters is a judgment call, and a capacity planner working with tight margins may set a lower one. The underlying relationship between MB and KB does not change; only the point at which you choose to care about the difference does.