Reuters Pictures API

Reuters Pictures API

The Reuters Pictures API is the most flexible and powerful option for searching and retrieving Reuters Pictures content. It collects items with precision and speed. It was designed and released under my supervision. As its Product Manager I specified its requirements, set the priorities, provided the voice of the customer, and oversaw its development, QA, and launch.

One challenge was building a system to serve our current and future needs while accommodating our legacy Pictures metadata conventions. Another was working with the third party image management company that hosts Reuters Pictures to fulfill our requirements within the constraints of their other customers. Some were already using a previous version so our new API features had to be backwards-compatible. The result makes the right compromises and provides great utility for customers, accomplished at the least expense.

Use Cases

The API supports a wide variety of applications. Here are just a few:

  1. A web console for browsing pictures from subscribed channels using filters and search terms, and downloading selected items.
  2. A production tool that algorithmically creates sideshows from pictures it collects matching specified criteria with no human effort.
  3. A system to automatically ingest pictures from subscribed channels in the desired formats as soon as Reuters publishes them.


Calls to the API are made via an HTTP request. Responses are delivered in XML, or JSON if specified by the request. Search queries may be simple or highly targeted. Items may match by the keywords assigned by Reuters, or more broadly anywhere within their associated text. Responses return a default set of item metadata, or the custom fields indicated to retrieve by the request. Links point where items may be downloaded in multiple sizes. Here are a few examples:

Get all items matching the keyword "tigers":

Get all items matching the text "tigers" that are of portrait orientation: AND Orientation:Portrait&token=ABC

Get all items matching a the text "tigers" or "lions" but not "Detroit": OR lions NOT Detroit&token=ABC

Get all items matching the text "Detroit Tigers"; retrieve each item's system id, title, long caption, and artist, sorted by date:"Detroit Tigers"&fields=SystemIdentifyer,Title,CaptionLong,Artist&sort=Date&token=ABC

...And these only scratch the surface.