Use the REST APIs to build a search UI

This is for:

Developer

Search integrations with Coveo typically rely on the Atomic or Headless libraries.

If, for some legitimate reason, you can’t use the Coveo Atomic or Headless libraries in your custom search integration with Coveo, you must ensure that your own search interface implementation uses the Search API and Usage Analytics Write API services correctly (that is, in a similar fashion to what the Coveo Atomic and Headless libraries do). Otherwise, Coveo Usage Analytics reports may become inconsistent/incoherent across your Coveo organization, the Coveo Machine Learning service may not be able to learn and function properly, and the overall end-user experience may be unsatisfactory.

The articles in this section provide guidelines for properly implementing your own search interface, assuming that you can’t use the Coveo Atomic or Headless libraries in your custom search integration with Coveo.

Note

These articles contain examples of search queries where the POST endpoint is used instead of the GET one, contrary to convention. We do so entirely for readability purposes, as the POST endpoint enables users to send a body parameter containing the required properties. If GET was used, those properties would have to be sent as distinct query parameters, which would make for very clumsy documentation.