IPTC vs XMP: Image Metadata Standards Explained
If you've worked with stock photography, news images, or digital asset management, you've encountered IPTC and XMP metadata. Both store editorial information about images (captions, keywords, copyright, credits) but they come from different eras and work differently.
IPTC-IIM: The Original Standard
IPTC-IIM (International Press Telecommunications Council – Information Interchange Model) was created in 1991 for the news industry. It was designed so photo agencies could transmit images with caption, credit, and keyword data over wire services.
IPTC-IIM is a compact binary format embedded in the image file (in JPEG, it lives in the APP13 marker). Key fields include:
- Headline: short title for the image
- Caption/Abstract: full description
- Keywords: searchable tags (can be multiple)
- By-line: photographer name
- Credit: who to credit
- Source: original provider
- Copyright Notice: rights statement
- City, State, Country: location of the scene (not GPS coordinates, but named locations)
- Date Created: editorial date (may differ from EXIF capture date)
- Special Instructions: embargo dates, usage restrictions
IPTC-IIM's strength is its simplicity: flat key-value pairs with fixed field sizes. Its limitation is that it's not extensible. You can't add custom fields, and it doesn't handle Unicode well in its original form.
XMP: The Modern Standard
XMP (Extensible Metadata Platform) was created by Adobe in 2001 as a universal, extensible metadata framework. Unlike IPTC-IIM's compact binary format, XMP is XML embedded directly in the file.
XMP uses namespaces to organize metadata into categories:
- dc: (Dublin Core) title, description, creator, subject (keywords), rights
- photoshop: city, country, credit, date created, color mode, ICC profile name
- xmp: creator tool, create date, modify date, rating
- xmpMM: document ID, instance ID, version history
- lr: Lightroom-specific data (develop settings, collections)
- crs: Camera Raw settings (exposure, white balance, etc.)
XMP's strength is extensibility. Anyone can define a new namespace with custom fields, which is why Adobe tools store their editing settings in XMP. The weakness is size: XML is verbose, and XMP packets can be several kilobytes.
How They Relate
XMP doesn't replace IPTC. It includes it. The IPTC organization defined a mapping called IPTC Core that puts all the classic IPTC-IIM fields into XMP namespaces. When you fill in the "Caption" field in Lightroom or Photoshop, it writes the value to both IPTC-IIM and XMP simultaneously.
This dual-write means most modern images contain the same editorial data in two places:
| Field | IPTC-IIM Location | XMP Location |
|---|---|---|
| Title | 2:5 Object Name | dc:title |
| Description | 2:120 Caption | dc:description |
| Keywords | 2:25 Keywords | dc:subject |
| Author | 2:80 By-line | dc:creator |
| Copyright | 2:116 Copyright Notice | dc:rights |
| City | 2:90 City | photoshop:City |
| Country | 2:101 Country Name | photoshop:Country |
Which One Should You Use?
Both, and you probably already are. Modern editing software (Lightroom, Photoshop, Capture One, Photo Mechanic) writes to both IPTC-IIM and XMP automatically. You don't need to choose.
If you're building a workflow or choosing what to read programmatically:
- Read XMP first: it's more complete, handles Unicode properly, and contains data that IPTC-IIM can't (like editing history, ratings, and custom fields)
- Fall back to IPTC-IIM: for older files that predate XMP, or files from systems that only write IPTC-IIM
- Write to both: for maximum compatibility. Some older DAM systems and wire services still only read IPTC-IIM
Conflicts Between IPTC and XMP
Because the same data lives in two places, they can get out of sync. If you edit the IPTC caption in one tool and the XMP caption in another, you have a conflict. Adobe's Metadata Working Group (MWG) published guidelines:
- When reading: prefer XMP over IPTC-IIM when they conflict
- When writing: update both simultaneously
- Lightroom and Photoshop follow these guidelines by default
In practice, conflicts are rare if you use one primary editing tool. They tend to happen when files pass through multiple systems in a production chain.
How EXIF Fits In
EXIF is the third piece of the puzzle. It stores technical data (camera, lens, exposure, GPS) rather than editorial data (caption, keywords, credit). EXIF is written by the camera automatically; IPTC and XMP are written by humans or software after capture.
Some fields overlap. Both EXIF and XMP can store the date taken, copyright, and artist. Again, XMP takes precedence per MWG guidelines.
View All Three in One Place
Our metadata viewer shows EXIF, IPTC, and XMP data side by side, decoded, organized, and readable. Drop any image to see exactly what metadata it contains.