Word and Character Counter

Why Are There Two Different Byte Counts? UTF-8 (3 Bytes) vs. EUC-KR (2 Bytes)

In this tool, the Korean character "가" is counted as 3 bytes in one field and 2 bytes in another. This isn't a bug; it's due to different standards. Misunderstanding this difference can lead to real-world costs in SMS messaging.

The "Bytes (UTF-8)" statistic uses the standard encoding for the modern web and databases, where one Korean character is 3 bytes. In contrast, the SMS/LMS gauge uses the EUC-KR standard common in Korean text messaging gateways, where a Korean character is 2 bytes. This means the same phrase "안녕하세요" can be either 15 bytes or 10 bytes, depending on the context.

The 90-Byte SMS Limit: Where Costs Can Spike

In Korea, the type and cost of a text message depend on its length. A Short Message Service (SMS) is limited to 90 bytes under the EUC-KR standard, which equates to 45 Korean characters. If you exceed this by even one byte, it's automatically sent as a Long Message Service (LMS), and the cost jumps by 2-3 times (the exact rate depends on your provider and contract).

This becomes a problem with bulk messaging. Sending a 91-byte message to 100,000 recipients incurs a significant extra cost that could have been avoided by trimming just two characters. When the gauge turns yellow (over 85%) or red, it's time to edit your message.

• An SMS is capped at 45 Korean characters or 90 English characters. For mixed-language text, the only way to be sure is to use this tool's byte gauge.
• A newline character also consumes 1 byte. It's not unheard of for a single blank line, added for readability, to push a message over the limit.
• For promotional messages, legally required prefixes like "(Ad)" and unsubscribe information also count towards the byte limit, reducing the space available for your main content.
• If you use placeholders (like for a customer's name), the final length will vary. You should calculate the length using the longest possible value. The LMS limit is 2,000 bytes.

Database Storage: How Many Characters Fit in VARCHAR(100)?

This is another common point of confusion. The answer depends on your database and its configuration.

• In MySQL 5.0+, VARCHAR(100) means 100 *characters*, not 100 bytes. With a `utf8mb4` character set, it can store 100 Korean characters.
• However, indexes have byte limits. Since `utf8mb4` characters can be up to 4 bytes, an index on a VARCHAR(255) column could require 1,020 bytes, exceeding older configuration limits like 767 bytes. This triggers the "Specified key was too long" error.
• In Oracle, VARCHAR2(100) defaults to 100 *bytes*, which holds only 33 Korean characters (at 3 bytes each). You must explicitly specify VARCHAR2(100 CHAR) to define the length in characters.
• In short: use the character count statistic when sizing MySQL columns, and the UTF-8 byte count statistic when calculating space for Oracle, indexes, and data transfer.

Gauging Length for Essays and Blog Posts

Here are some common guidelines for writing to a specific length. These aren't strict rules, but practical rules of thumb.

• In Korea, traditional manuscript paper (*wongoji*) holds 200 characters per sheet, including spaces. An assignment for "5 sheets" means 1,000 characters. You can estimate this by dividing the "Characters (with spaces)" count by 200.
• Cover letter prompts often have limits of 500-1,000 characters, usually including spaces. Always check the specific guidelines on the application portal.
• For SEO, a common benchmark for blog posts is over 1,000 characters (excluding spaces). The character count itself isn't a ranking factor, but rather a baseline to filter out thin content.
• The 32-character blog title gauge is an approximation for what won't get cut off in search engine results. Since Google truncates titles by pixel width, not character count, this is an estimate.
• Reading time is calculated at 500 characters per minute (an approximation of the average silent reading speed for Korean adults). You can use this for displaying an "Estimated read time," but be aware that average reading speeds vary by language.

Word Count is Only a Reference for Korean

The "Words" statistic is calculated by splitting the text by spaces. While this works well for languages like English, it's less precise for Korean. For example, the phrase "밥을 먹었다" (I ate a meal) is counted as two words, but is composed of multiple meaningful units (morphemes), and the word count changes depending on spacing. This is why character count, not word count, is the standard convention for measuring text length in Korean.

The "Top 10 Most Frequent Words" list has the same limitation. Because grammatical particles are attached to nouns, "고객이" (customer, as a subject) and "고객을" (customer, as an object) are counted as two different words. It's a simple frequency count without morphological analysis, but it's still useful for spotting repetitive phrasing at a glance.

• Common, low-meaning tokens like particles and conjunctions (e.g., and, but, is, the) are excluded from the list to highlight more significant words. Words shorter than two characters are also ignored.
• If the same word appears multiple times in the top rankings of your cover letter, it's a signal that you should diversify your phrasing.