Best practice - CSV downloads

In CSV download processes, Pacific Islander language characters (e.g., Māori, Samoan, Tongan, Fijian, Hawaiian) are usually handled through character encoding, and the key factor is whether the CSV is encoded in UTF-8.

If the encoding is handled correctly, characters like macrons or diacritics will export and import properly. If not, they often become garbled (mojibake).

1. Common Characters in Pacific Languages

Examples include macrons and special letters:

Language

Example

Character

These characters are Unicode characters, not basic ASCII.

2. Correct Handling (UTF-8)

Best practice for CSV exports today is:

UTF-8 encoding

Example CSV row:

Name,City
Tāmaki Makaurau,Auckland
Hawaiʻi,Honolulu

UTF-8 correctly preserves:

• macrons (ā ē ī ō ū)

• glottal stops (ʻ)

• accented characters

Most modern systems support UTF-8:

• web apps

• AWS pipelines

• Postgres

• Excel (recent versions)

• Google Sheets

3. Where Problems Usually Occur

Issues typically appear when CSV files are opened in older Excel workflows.

Excel historically assumes:

Windows-1252 encoding

Instead of UTF-8.

Result:

Intended

Garbled

This is not data corruption, just incorrect decoding.

4. Standard CSV Export Best Practice

Good export pipelines do one of the following:

Option 1 — UTF-8 with BOM (most Excel-friendly)

UTF-8 + BOM

This helps Excel detect encoding.

Example first bytes:

EF BB BF

Many SaaS apps add this automatically.

Option 2 — UTF-8 without BOM

Standard for APIs and data pipelines.

Works fine in:

• Google Sheets

• modern Excel import

• databases

• programming languages

5. Example Export Pipeline

Typical SaaS CSV generation:

Database (UTF-8)
        ↓
application layer
        ↓
CSV serializer
        ↓
UTF-8 encoded file
        ↓
download

Characters remain intact as long as encoding stays UTF-8.

Excel and google sheets work differently

Google Sheets actually always exports CSV as UTF-8. There is no option to choose encoding, which is why you don’t see a “Save as UTF-8” setting.

How Google Sheets CSV export works

When you download:

File → Download → Comma-separated values (.csv)

Google automatically produces:

UTF-8 encoded CSV

So characters like:

• ā ē ī ō ū

• ʻ (okina)

• é ñ ü

are preserved correctly in the file.

Where confusion usually happens

The issue normally appears after download, when the CSV is opened in Excel.

Older Excel behaviour:

CSV opened directly → assumes Windows-1252 encoding

Result:

Correct

Excel shows

The CSV itself is fine — Excel is interpreting it incorrectly.

Correct way to open UTF-8 CSV in Excel

Instead of double-clicking the file:

1. Open Excel

2. Go to Data

3. From Text / CSV

4. Select the file

5. Set File Origin → UTF-8

Then characters display correctly.