Import a Photo Inventory CSV Without Losing Names or IDs
By ImgHeal · Updated
Import a photo inventory with explicit text columns and the correct character encoding. Opening a CSV directly can change identifiers and dates before you notice. Test a few difficult records first, then compare the exported values with the original text file before trusting the full inventory.
Explore this topic: Organize a photo archive

Build a small import test
Make a copy of the CSV and choose records that reveal likely mistakes. Include an ID beginning with zero, a name in the collection's actual language, a caption containing a comma, and an uncertain date. Use made-up records for software tests when you do not need a real person's information. Keep the received CSV unchanged as the comparison source.
The example contains one deliberately ambiguous date label. It belongs in a text field because it describes uncertainty rather than a date you can calculate with. A leading-zero item ID is also text. Neither should be converted merely because a spreadsheet recognizes part of its value as a number.
photo_id,name,date_note,caption
0007,Meera,about 1940,"At home, beside the gate"
0008,José,1940-1945,Unidentified gatheringChoose encoding and column types before loading
In supported Excel editions, start with Data, Get Data, From Text/CSV. Inspect the delimiter and encoding preview. Microsoft documents this import route for UTF-8 files that do not open correctly by default. Verify a known accented or non-Latin name in the preview rather than assuming a correctly displayed English heading proves that every row decoded properly.
Set identifier and date-note columns to Text before automatic conversion is accepted. If Power Query adds a Changed Type step that already converted those fields, return to the import steps and correct or remove that conversion before loading. Setting a damaged value to Text afterward does not recreate digits or spelling that were already lost.
Check values, not just their appearance
A display format can make the number 7 look like 0007 while leaving its stored value numeric. Test what is actually exported. Save a separate UTF-8 CSV, open that output in a text editor, and compare the difficult records, column count, commas inside captions, and row count. Preserve the editable workbook separately if it contains formulas, formatting, or multiple sheets that CSV cannot represent.
| Test value | Expected result |
|---|---|
| 0007 | Four characters, including all leading zeros |
| José | Accented name unchanged |
| 1940-1945 | Text range, not an invented exact date |
| At home, beside the gate | One caption field despite its comma |
Link restorations after the inventory survives export
Use the stable photo ID to connect the source scan and any ImgHeal viewing copy. Add a derivative filename or editing status without replacing the original caption. Ask another family member to locate one source and its restored copy using the exported inventory. This tests the handoff people will use, rather than only the workbook on your computer.
Inventory import checks
- Keep the received CSV unchanged.
- Import IDs and uncertain dates as Text.
- Verify multilingual text and quoted captions.
- Inspect a fresh export in a text editor.
Sources and further reading
- Microsoft: Keep leading zeros and large numbersExplains automatic conversion, text values, and the limitations of display formats.
- Microsoft: Open UTF-8 CSV filesDocuments encoding-aware import routes.
Questions about this restoration task
Can formatting the column restore lost leading zeros?
It can change appearance, but cannot reliably recover the original ID. Return to the untouched CSV and import the column as text. Do not guess how many zeros were present.
Continue with a related guide
Try a restoration on a copy of your photo
Keep the original scan. Upload a separate digital copy, review ImgHeal’s generated result, and decide whether it suits your needs.
Open ImgHeal