{"site":{"name":"Koji","description":"AI-native customer research platform that helps teams conduct, analyze, and synthesize customer interviews at scale.","url":"https://www.koji.so","contentTypes":["blog","documentation"],"lastUpdated":"2026-11-22T00:07:56.576Z"},"content":[{"type":"documentation","id":"1ff4d3b4-58b8-45aa-8fa4-c6c93a55fc1f","slug":"segment-research-integration","title":"Segment + Koji: Trigger AI Interviews from CDP Audiences and Sync Insights Back as Traits","url":"https://www.koji.so/docs/segment-research-integration","summary":"The Koji + Segment integration is bidirectional and requires no native connector. From Segment into Koji: a Segment Audience (Engage/Personas) or a forwarded track event starts a Koji AI-moderated interview for each matching user via Zapier, a Segment Function, or reverse ETL (Hightouch/Census), with the Segment userId as respondent.external_id. Koji can also email invitations itself to an audience imported into Participants, on Insights, Interviews and Enterprise. From Koji back into Segment: a scheduled worker polls the Koji read endpoint and, when an interview comes back completed with its analysis (summary, key insights, sentiment), calls Segment's HTTP Tracking API to post a `Koji Interview Completed` track event and identify traits like last_interview_sentiment, which Segment fans out to the warehouse, Braze, Customer.io, Mixpanel, and dashboards. Outbound webhooks are not available yet, so the return leg polls. The headless API and the MCP connector are included on the Interviews and Enterprise plans. Design studies with Koji's six structured question types so answers return to Segment as clean, segmentable traits.","content":"# Segment + Koji: Turn CDP Audiences Into AI Interviews — and Pipe the Answers Back Into Every Tool\n\n**Answer first:** Segment is the system of record for *who* your users are and *what* they do. Koji's AI-moderated interviews answer the question a CDP structurally cannot: *why*. The integration runs both directions. **Into Koji:** a Segment Audience or a forwarded Segment event triggers a Koji AI interview to exactly those people — trial stallers, milestone-hitters, freshly-churned accounts. **Back into Segment:** a small worker polls Koji for finished interviews and sends a `Koji Interview Completed` track call that updates user traits like `last_interview_sentiment` and `last_interview_themes`, which Segment then fans out to your warehouse, email platform, and dashboards automatically. There is no one-click connector to install, and no outbound webhooks either: Koji exposes a headless API that a Segment Function, Zapier, n8n or Make can call in both directions. A working v1 takes about 30 minutes.\n\nIf you have ever built a Segment Audience called \"activated but silent for 14 days\" and wished you could just *ask them* what happened, this is the integration that turns that audience into a stack of transcripts by Friday.\n\n## Why connect Segment and Koji\n\nA CDP is superb at behavioral truth and blind to motivation. Teams usually patch that blindness with low-signal tactics: a quarterly NPS blast, a Slack message to CS asking for \"vibes,\" or five calendar-negotiated interviews a month that sample a rounding error of the cohort. Koji removes the scaling ceiling — its AI moderator runs voice or text interviews around the clock, with no scheduling and no human moderator — but the real unlock is **precision targeting**. You do not want to interview \"users.\" You want to interview the 212 people in one specific Segment Audience. Segment already computes those audiences continuously; the integration simply hands each new member to Koji.\n\nBecause Koji's AI asks adaptive follow-up questions instead of running a fixed script, one triggered interview produces the kind of \"why\" a survey never reaches — and every answer flows back into the same Segment identity you started from. Platforms like Koji make the depth of a moderated interview available at survey scale.\n\n## What flows in each direction\n\n### Segment → Koji (trigger interviews from behavior)\n- A user enters a Segment **Audience** (Engage/Personas) or fires a qualifying **track** event.\n- Segment forwards that to Koji through one of three transports (below).\n- Your handler starts a Koji interview for that person, with the Segment `userId` as `respondent.external_id`. Koji returns an `interview_id` and a session token, and the respondent takes the interview in an in-product embed or your own chat surface. To invite people by email instead, export the audience as a CSV, import it into Koji Participants and let Koji email each person their own interview link from the study's Self-recruit tab (Insights, Interviews and Enterprise plans), with sent, opened and started tracking.\n\n### Koji → Segment (insights become traits and events)\n- Koji has no outbound webhooks, so a scheduled worker polls `GET /api/v1/interviews/{id}` for the ids it is waiting on and acts when `status` is `completed` and `analysis` is present.\n- The worker calls **Segment's HTTP Tracking API**: a `track` event named `Koji Interview Completed` (properties: summary, key insights, sentiment, duration) plus an `identify` call setting durable traits.\n- Segment fans those traits to every destination — Snowflake/BigQuery, Braze, Customer.io, Mixpanel, your BI dashboards — with zero extra wiring.\n\n## Three ways to wire it up\n\n**1. No-code with Zapier (about 30 minutes).** Use Segment's Zapier destination (or an Audience webhook) as the trigger and a Zapier HTTP step calling Koji's Start Interview endpoint as the effect. There is no Koji Zapier app. For the return trip, run a scheduled Zap that polls Koji's read endpoint and uses Segment's Zapier action to send `track`/`identify`. Best for a first test — see [Connect Koji to Zapier](/docs/zapier-research-automation).\n\n**2. Segment Functions worker (about 1 hour).** Write a small scheduled function that polls Koji's read endpoint and calls Segment's Tracking API, plus an **Audience webhook** (or Source Function) that calls Koji's [headless API](/docs/headless-api-overview) to start interviews. This is the clean production path and keeps everything inside Segment.\n\n**3. Reverse ETL / warehouse-first (production-grade).** If your audiences live in the warehouse, sync them to a webhook with Hightouch or Census and call Koji's [Start Interview endpoint](/docs/starting-interviews-via-api). Koji insights land back in the warehouse as a modeled table you can join to any metric.\n\n## Identity resolution: matching interviews to the right user\n\nTwo clean patterns:\n- **ID-keyed (recommended):** pass your Segment `userId` as `respondent.external_id` when you start the interview. The read endpoint returns it unchanged under `respondent.external_id`, so your worker can `identify` the exact profile, even for pseudonymous users.\n- **Extra context:** anything else you want stored with the respondent, such as plan or signup date, goes in `respondent.metadata`. Keep email addresses out of it unless you need them.\n\nFor sensitive studies, run the interview in **anonymous mode**: no email is collected, and you simply skip the Segment write-back.\n\n## Design the interview so the data is actually joinable\n\nThe reason Segment + Koji beats a fancy survey link is Koji's **structured questions**. Every study can mix six question types — `open_ended`, `scale`, `single_choice`, `multiple_choice`, `ranking`, and `yes_no` — and each produces a typed, aggregatable value alongside the conversational transcript. A `scale` NPS item or a `single_choice` \"primary reason\" becomes a clean property you can write straight back to Segment as a trait and segment your entire base by. See the [structured questions guide](/docs/structured-questions-guide) to design questions that analyze themselves.\n\n## High-value Segment audiences to point at Koji\n- **Activation drop-offs** — signed up, never reached the aha event (feed [onboarding drop-off research](/docs/onboarding-drop-off-research-guide)).\n- **Expansion signals** — power users still on a low tier (feed [account expansion research](/docs/expansion-upsell-research-guide)).\n- **Silent churn risk** — usage decayed 50% month-over-month.\n- **Just-churned** — subscription cancelled in the last 24 hours (route to [cancel-flow exit interviews](/docs/cancel-flow-exit-interview)).\n- **New power users** — crossed a usage milestone; capture testimonials while enthusiasm is high.\n\n## Example: closing the trial-conversion loop\n1. Segment Audience: `trial_started` AND NOT `subscription_created` AND `days_since_signup >= 10`.\n2. Audience entry triggers a 4-question Koji AI interview (one `single_choice` on the main blocker, two `open_ended` probes, one `scale` on likelihood to pay).\n3. Interview completes: `Koji Interview Completed` lands in Segment with `theme = \"SSO missing\"`, `sentiment = \"negative\"`.\n4. Segment routes it to Snowflake (for the PM's trial-conversion model) and to Customer.io (which enrolls those SSO-blocked users in a tailored win-back sequence).\n\nYou now segment trial conversion by *stated reason*, not guesswork — exactly what [free trial feedback research](/docs/free-trial-feedback-guide) is for, running continuously.\n\n## Plans and limits\nThe headless API, API keys and the MCP connector are included on the **Interviews** and **Enterprise** plans. On other plans the API answers with a 403 asking for an Interviews plan or higher. Recruitment emails sent by Koji are available on Insights, Interviews and Enterprise. Outbound webhooks are not available yet, so the return leg polls the read endpoint. Zapier and CSV import cover the no-code version without writing code. Interviews start as low as €1 per qualified interview, and €3 per qualified voice interview. You pay only for the interviews your study actually uses.\n\n## The interview payload, field by field\n\nWhen you read a finished interview, Koji returns a JSON object your worker maps onto Segment. The fields you will use most:\n- `interview_id` and `status`: which interview this is and whether it has finished.\n- `respondent.external_id`: the identity key you passed in, returned unchanged for joining.\n- `analysis.summary`: a short written summary of the conversation.\n- `analysis.key_insights`: the main points the respondent made.\n- `analysis.sentiment`: an overall label (`positive`, `neutral`, `negative` or `mixed`).\n- `stats`: message counts and duration in seconds.\n- `transcript`: the full conversation, message by message.\n\n`analysis` stays `null` until the interview is completed and its analysis has finished, so keep polling until it is present.\n\nBecause you decide which of these to forward, you keep tight control over what ever lands in your CDP.\n\n## Common pitfalls (and how to avoid them)\n- **Interviewing everyone at once.** A CDP makes it tempting to fire at a 10,000-person audience. Start with a tight, high-intent slice (a few hundred) so you can read every transcript and refine the questions before scaling.\n- **Skipping identity mapping.** If you forward insights without a stable `respondent.external_id`, they will not join back to the right profile. Decide your key — email or ID — before you build the return leg.\n- **Writing raw transcripts to Segment.** Keep the CDP clean and privacy-safe by forwarding only structured insights; leave transcripts in Koji.\n- **No frequency cap.** Suppress users who were interviewed in the last N days so a busy audience does not receive repeat invitations.\n\n## What to expect: week one vs. quarter one\n\nIn week one you will typically wire a single audience-to-interview flow and watch the first transcripts land. By the end of a quarter, most teams run three to five always-on flows — activation, expansion, churn, and testimonial capture — each feeding themed traits back into Segment. At that point qualitative \"why\" is no longer a quarterly project; it is a live property on every user, refreshed continuously and available to every downstream tool the moment it is created.\n\n## Related Resources\n- [Research Automation](/docs/research-automation-webhooks)\n- [Connect Koji to Zapier](/docs/zapier-research-automation)\n- [Starting Interviews via API](/docs/starting-interviews-via-api)\n- [Structured Questions in AI Interviews](/docs/structured-questions-guide)\n- [Mixpanel + Koji](/docs/mixpanel-research-integration)\n- [Amplitude + Koji](/docs/amplitude-research-integration)\n- [Understanding Themes & Patterns](/docs/understanding-themes-patterns)","category":"API Reference","lastModified":"2026-10-29T05:10:23.461656+00:00","metaTitle":"Segment + Koji Integration: AI Interviews From CDP Audiences","metaDescription":"Trigger Koji AI interviews from Segment Audiences and events, then sync summaries, key insights and sentiment back into Segment as track events and user traits.","keywords":["segment koji integration","segment cdp research","trigger interviews from segment audience","customer data platform research","segment webhook interviews","ai interviews from segment","segment reverse etl research"],"aiSummary":"The Koji + Segment integration is bidirectional and requires no native connector. From Segment into Koji: a Segment Audience (Engage/Personas) or a forwarded track event starts a Koji AI-moderated interview for each matching user via Zapier, a Segment Function, or reverse ETL (Hightouch/Census), with the Segment userId as respondent.external_id. Koji can also email invitations itself to an audience imported into Participants, on Insights, Interviews and Enterprise. From Koji back into Segment: a scheduled worker polls the Koji read endpoint and, when an interview comes back completed with its analysis (summary, key insights, sentiment), calls Segment's HTTP Tracking API to post a `Koji Interview Completed` track event and identify traits like last_interview_sentiment, which Segment fans out to the warehouse, Braze, Customer.io, Mixpanel, and dashboards. Outbound webhooks are not available yet, so the return leg polls. The headless API and the MCP connector are included on the Interviews and Enterprise plans. Design studies with Koji's six structured question types so answers return to Segment as clean, segmentable traits.","aiPrerequisites":["A Koji account (API access is included on every plan, including free)","A Segment workspace with Engage/Audiences or event access","Familiarity with webhooks or Zapier"],"aiLearningOutcomes":["Trigger Koji AI interviews from a Segment Audience or event","Send interview insights back into Segment as track events and traits","Resolve identity between Segment and Koji with email or respondent_id","Choose between Zapier, Segment Functions, and reverse ETL setups"],"aiDifficulty":"intermediate","aiEstimatedTime":"12 min read"}],"pagination":{"total":1,"returned":1,"offset":0}}