Mailchimp + Koji: Turn Email Journeys and Audiences into AI Customer Interviews

Connect Mailchimp to Koji and every Customer Journey becomes a doorway to a real conversation. Instead of dropping a static one-question survey into a broadcast email, you trigger a Koji AI interview — voice or text — the moment a contact hits a journey point, then write the themes and structured scores back onto their Mailchimp profile so your next segment is built on why, not just opens and clicks.

Mailchimp is where marketers already orchestrate the customer relationship: welcome series, post-purchase flows, win-back campaigns, and audience segments. Koji is the AI-native research layer that turns any of those moments into a moderated interview at scale — no recruiter, no scheduling, no moderator sitting in on calls. This guide shows you exactly how the two connect and which journeys are worth wiring up first.

What data flows between Mailchimp and Koji

The integration is bidirectional:

• Mailchimp → Koji (trigger): When a contact enters a Customer Journey, gets a tag, joins a segment, or completes an e-commerce action, Mailchimp fires an event that starts a personalized Koji interview and emails or texts them a unique link.
• Koji → Mailchimp (write-back): When the interview finishes, Koji writes an interview status, a short AI summary, the top themes, and typed scores from your structured questions back onto the Mailchimp contact — as merge fields, tags, or segment membership you can immediately target.

Because Koji's outputs come from structured questions with typed values (an NPS scale, a single-choice churn reason, a yes/no repurchase intent), they land in Mailchimp as clean, segmentable data — not a blob of free text you have to read by hand.

Three ways to connect

1. No-code with Zapier or Make (recommended to start). Mailchimp exposes triggers like New Subscriber, New or Updated Subscriber, and Tag Added in both Zapier and Make. Wire one of those to Koji's "start interview" action. On the return trip, use Koji's "interview completed" trigger to run a Mailchimp Add/Update Subscriber, Add Tag, or Update Merge Field action. Most teams have a working loop in under an hour.

2. Real-time with webhooks. Mailchimp's per-audience webhooks push real-time POSTs for subscribe, profile, unsubscribe, and campaign events. Point them at a small handler that calls Koji's API to launch an interview. When Koji fires its own completion webhook, your handler calls the Mailchimp Marketing API to patch the contact's merge fields. This is the lowest-latency path and the one to use for high-volume flows.

3. AI-driven with Koji's MCP and API. Koji ships a Model Context Protocol integration (15 tools) plus a developer API, so you can drive the whole workflow from Claude or your own code: create a study, import respondents from a Mailchimp segment export, launch interviews, and pull structured results back — all programmatically.

The best Mailchimp journeys to add interviews to

You don't need to interview everyone. Add Koji to the three moments where email alone under-delivers:

Post-purchase voice-of-customer. Add a journey point roughly 10–14 days after a purchase (or after the "order delivered" event from a connected store). A Koji interview captures not just a satisfaction rating but the hesitation they felt before buying, the alternative they almost chose, and the feature they wish existed. That is roadmap and messaging fuel a star-rating never gives you.

Win-back and lapsed-subscriber exit interviews. Mailchimp's re-engagement and churn-risk journeys are the perfect trigger. Instead of one more discount email, ask the real question: what changed? Koji's AI follow-ups dig past "too expensive" to the actual job the product stopped doing — the difference between a guess and a fix.

VIP and repeat-buyer roadmap research. Tag your highest-LTV segment and route them into a short Koji study that ranks the features they want next. Your most valuable customers get heard first, and you get prioritized input from the people whose retention matters most.

What syncs back to Mailchimp

After each completed interview, Koji can write to the contact:

• Interview status (e.g. koji_status = completed) as a merge field or tag
• A short AI summary of the conversation
• Top themes surfaced across the interview
• Structured question scores — an NPS value, a CSAT scale point, a single-choice reason, a yes/no signal — each a typed field you can segment or trigger on
• A link to the full transcript and Koji report

That means your next Mailchimp segment can be "detractors who cited onboarding" or "repeat buyers who ranked integrations #1" — audiences you simply cannot build from opens and clicks.

Why a conversation beats a one-question email survey

Timing and relevance drive survey response far more than the survey tool itself — feedback requests sent close to a real experience consistently outperform generic embedded email surveys, which often languish in the low single digits to mid-teens. But response rate is only half the story. A rating tells you the score; it never tells you the reason.

Koji closes that gap. Its AI interviewer asks adaptive follow-up questions in voice or text, so a "6 out of 10" becomes a two-minute explanation of exactly what went wrong. And it supports all six structured question types in a single conversation — open_ended, scale, single_choice, multiple_choice, ranking, and yes_no — so you capture clean quantitative signals and the narrative behind them at once. See the structured questions guide for how to combine them.

Koji's quality gate keeps the data honest: each session is scored for effort and coherence, so one-word or nonsense responses are flagged, excluded, and never consume credits. Only genuine conversations flow back to your Mailchimp audience. Koji's free tier includes 10 one-time credits to test the loop; paid plans start at €29/mo (Insights) and €79/mo (Interviews), with text interviews costing 1 credit and voice 3.

Mapping Mailchimp fields to Koji structured questions

A little planning up front keeps the write-back clean. Decide which merge fields will hold Koji's results before you build the loop, because Mailchimp merge fields are defined per audience and referenced by their merge tag (for example *|KOJINPS|*).

A typical mapping looks like this:

• KOJISTAT (text) ← Koji interview status (invited, started, completed)
• KOJINPS (number) ← the scale question score, so you can segment detractors vs promoters
• KOJIREASON (text) ← the single_choice churn or hesitation reason
• KOJITHEME (text) ← the top theme Koji surfaced
• KOJILINK (text) ← a link to the full transcript and report

Create those merge fields once, then have your Zapier, Make, or webhook step patch them on interview completion. Because the scale and single_choice values arrive as typed data, you can build a live segment such as "KOJINPS ≤ 6 AND KOJIREASON = onboarding" and drop those contacts straight into a recovery journey.

Common setup mistakes to avoid

Interviewing on the broadcast, not the moment. Blasting an interview link to a whole audience gets the same fatigue as a mass survey. Trigger from a specific journey point — a purchase, a churn-risk signal, a tag — so the ask always lands in context.

Forgetting the throttle. Add a wait/delay and a suppression tag so the same contact is not invited to two interviews in the same month. A simple "interviewed in last 90 days" segment used as an exclusion keeps the experience respectful.

Storing free text instead of typed scores. The whole advantage of Koji is that its structured questions return values you can segment on. Map those to number and single-choice merge fields, not one catch-all text field, and your Mailchimp audiences stay queryable.

Start with one journey — post-purchase is the easiest win — prove the loop end to end, then expand to win-back and VIP research once the write-back is landing cleanly on your contacts.

Related Resources

• Structured Questions Guide — the 6 question types that make interview data segmentable
• Klaviyo + Koji Integration — the same pattern for Klaviyo flows and segments
• Zapier Research Automation — build the no-code loop step by step
• Make Research Automation — the Make.com alternative
• Webhook Setup — the real-time, low-latency path
• Post-Purchase Survey Guide — what to ask after a sale
• How the Quality Gate Works — why low-effort responses never pollute your data