++ Intelligence Brief and Intelligence Stream Field Definitions

HubSpot, Salesforce, Pipedrive and Slack use a default, pre-built integration, so there is no field mapping required there. If you are connecting through the Zapier App, Zapier Webhook or a Standard Webhook, you can map the following Bread & Butter data points directly to your own fields.

In short: this article lists every field available in the Intelligence Brief and Intelligence Stream. HubSpot, Salesforce, Pipedrive and Slack use a default, pre-built integration, so no field mapping is needed there; field mapping only applies when you are connecting through the Zapier App, Zapier Webhook or a Standard Webhook. The Brief carries identity, firmographic, attribution and written sales-context fields plus the latest Nurture draft, all at the top level of the payload. The Stream carries everything in the Brief plus the full raw material behind it: engagement detail, top pages, complete session-by-session history and the full Nurture draft history. Jump to a section below, or read on for details on when each field populates.

Two columns below tell you where each field is available. The Intelligence Brief carries the summary; the Intelligence Stream carries the summary plus all the raw material behind it. The Stream is a strict superset of the Brief: every Brief field appears in the Stream under the same name. The Stream is available only through Zapier and Webhooks, because the payload is larger than standard CRM field mapping can absorb.

Field names below are given exactly as they appear in the payload. Timestamps are ISO 8601 with three fractional digits and a trailing Z, for example 2026-09-18T14:32:07.412Z . A value that is absent arrives as null , never as an empty string, so your field mappings never break.

Session

User Profile (user )

These fields capture the lead's core identity, professional background and current intent level.

Company (company )

These fields deliver deep firmographic enrichment based on the lead's employer.

First Attribution (first_attribution )

These fields capture the origin data from the lead's very first visit to your website.

Click identifiers are passed through exactly as they arrived in the URL, capitalisation included, so a value you receive can be matched against your ad platform without transformation.

Last Attribution (last_attribution )

These fields capture data from the lead's most recent session, the visit that just closed. It includes the exact same location, device and UTM fields as first_attribution , letting you see what specifically brought a returning lead back to your site today.

Written Sales Context

Six AI-generated fields designed for immediate outreach, written to be scanned in about ten seconds. They sit at the top level of the payload alongside user  and company ; there is no wrapper object around them.

Nurture (nurture_draft , nurture_sent , nurture_drafts )

Nurture AI generates a draft email each time a lead returns. The Brief carries the current draft and every email actually sent; the Stream adds the full history of every draft ever generated.

Top Engaged Pages (top_engaged_pages )

Stream only. The pages that have held this lead's attention most across all sessions, as structured data, ordered by total engagement score with the highest first. It sits at the top level of the payload beside sessions . The Brief describes the same thing in prose through user_journey_summary  and latest_session_summary .

Session History (sessions )

Stream only. An array containing every session the lead has ever had on your site, not just the most recent one. This is the raw material behind the summaries above, and it is what makes the Stream suited to data warehouses and BI tools rather than CRM fields.

Sessions are ordered newest first. There is no cap on the number of sessions, but if a lead's history would push the payload past roughly 5 MB, older sessions are dropped, sessions_truncated  is set to true  and session_count  still reports the real total. In practice this only happens to bot traffic; a real lead's entire history is a few tens of kilobytes.

Event types

sessions[].events[]  is a closed list. Every entry carries a type and a timestamp, and your own events also carry the name you gave them in name .

Page views and outbound clicks are not events. Pages live in sessions[].pages[]  and outbound clicks in sessions[].exit_destinations[] , so you never have to filter them back out of the event array.

Exit destinations

sessions[].exit_destinations[]  records the outbound links the lead clicked during that session, which is often where the real intent shows: a booking tool, a review site, a comparison page. Each entry carries a url  and a timestamp , and nothing else.

When a session contained no outbound click the field is null  rather than an empty array. That is the normal value for most sessions rather than a sign of anything missing. Exit destinations are Stream only; the Brief carries no exit destination field.

When Fields Are Populated

Not every field arrives on every payload.

UTM fields and click identifiers are always present in every attribution object. They are null  when a lead arrives directly or through organic search, since there is no campaign to report, so your field mappings never break.

Enrichment fields under user  and company  populate once enrichment has run against that lead. Newly captured leads may arrive before enrichment completes.

focused_score  is null  until the lead has been scored, so you can tell "scored zero" apart from "not scored yet". previous_focused_score  is null  until the score has moved at least once.

number_of_employees  is null  when the figure was found only as a range, and number_of_employees_text  carries the range instead, so a warehouse can filter on the number and a CRM field can display the text.

real_time_icebreaker  requires a resolvable location. Leads on VPNs or corporate proxies may not receive one.

last_attribution  matches first_attribution  exactly on a lead's first session, since it is the same visit.

lead_stage  carries the stage's current label as shown in your dashboard, so a renamed stage changes the value that arrives. Warehouses should join on lead_stage_rank  in the Stream instead.

nurture_draft  is absent until Nurture AI has generated a draft for the lead.

Every field is Bread & Butter's own. Nothing in either payload passes through a third-party data vendor's raw response or field names, so what you map today stays stable if a source changes behind the scenes.

Related

• The Intelligence Brief: Living Context in the Tools Your Team Already Uses
• The Intelligence Stream: The Complete Record Behind Every Summary
• Integrations and Automation: what you can send, where, and when