Connection summary
Data handling
Intercom’s schema endpoint does not list every custom attribute in use. Rippit samples recent conversations to find the rest and offers them in the field picker, so you can add attributes the API does not advertise. Intercom’s own CX Score is one example. Attributes created after you connect are found the same way. Only fields you select become columns. During setup, the picker lists only attributes that have a value in the sampled conversations.
The field picker also offers company attributes, listed with a “Company” prefix, and conversation fields Intercom nests outside custom attributes, such as SLA name and status, handling time, and AI Agent fields.
API endpoints
Authentication
Conversations and attributes
Comments
Conversations are read in two phases, which is why
/conversations/search and /conversations/<id> both appear.
Discovery. /conversations/search asks which conversations changed in a given time window, 150 per page. It returns conversation records with a message count but not the messages themselves. It uses POST because that is Intercom’s search API contract; it returns data and changes nothing. Search results are used only for their IDs. Rippit trails Intercom by two hours to let updates settle, and re-sweeps the last 24 hours once a day.
Content. Each conversation the search names is then fetched individually at /conversations/<id>, 5 concurrent. This is the call that returns conversation_parts, the actual message history. Custom attribute discovery fetches a small sample the same way, 10 concurrent, at most every 6 hours.
The workflow: one search call per 150 changed conversations in each window, plus one GET per changed conversation and one per contact on it.
Contacts are fetched by ID only, never listed or searched. Rippit takes the contact IDs on each changed conversation that search returns and retrieves those records one at a time, 10 concurrent. A contact is read again whenever one of its conversations changes. There is no sweep of your contact base, so a contact who never appears on a synced conversation is never read.
Attachments are requested without credentials first. The access token is sent only when the file is hosted on an Intercom domain. Inline images are always fetched without credentials.
Sync behavior
Rippit selects conversations by last update, not creation. A conversation created before your backfill start but updated after it is synced with its original created date, so your table can hold conversations older than the backfill window you set.