monday.com + Koji: Turn Board Automations into AI Customer Interviews

Connect monday.com to Koji and a status change becomes a customer conversation. When an item moves to "Done," a deal reaches a stage in monday CRM, or a lead is marked qualified, a monday automation fires a webhook that starts a Koji AI interview — voice or text — then writes the themes and structured scores back into the item's columns. The feedback that usually lives in someone's head lands directly on the board where the work happens.

monday.com is the Work OS where teams run projects, pipelines, and roadmaps on boards, groups, items, and status columns. Koji is the AI-native research layer that turns any board event into a moderated customer interview at scale — no recruiter, no scheduling, no moderator. Together they close the loop between "we shipped it" and "here is what customers actually think."

What data flows between monday and Koji

The integration runs both directions:

• monday → Koji (trigger): When a status column changes, an item is created, or a monday CRM deal moves stage, a monday automation posts a webhook that launches a Koji interview and sends the contact a unique link.
• Koji → monday (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 into the item — as column values and an Update (comment) on the item.

Because Koji's outputs come from structured questions with typed values — a scale satisfaction score, a single_choice reason, a yes/no signal — they populate monday Status, Number, and Dropdown columns cleanly, so you can sort, filter, and build dashboards on the customer's answer, not just the task's progress.

Three ways to connect

1. No-code with monday automations + webhooks (recommended). monday's automation recipes follow a simple "when trigger, do action" model — "when status changes to Interview, send a webhook." Point that webhook at Koji's API to launch the interview. On the return trip, Koji's completion webhook calls the monday API to set column values and post an Update. This keeps the whole loop inside monday's native automation builder.

2. No-code with Zapier or Make. monday exposes triggers like New Item in Board, Specific Column Value Changed, and New Update in both Zapier and Make, plus actions like Create Item, Change Column Value, and Create Update. Wire a column-change trigger to Koji's start-interview step, and write results back with a change-column-value action. One practical note: writing to a monday column requires that column's unique ID, which is specific to each board.

3. AI-driven with Koji's MCP and API. Koji ships a Model Context Protocol integration (15 tools) plus a developer API, and monday offers a GraphQL API. Read items from a board, feed the contacts into a Koji study, launch interviews, and write structured results back into columns — all driven from Claude or your own code.

The best monday workflows to add interviews to

monday CRM: won and lost deals. In monday CRM, deals move through stages on a board. Trigger a Koji interview when a deal is marked Won or Lost to capture the buyer's real reason — who they chose, what tipped it — and write a verified reason back to the deal's dropdown column. Your pipeline dashboard now groups outcomes by customer truth, not a rep's guess.

Project delivery: post-launch feedback. When a feature or client project item moves to "Done" or "Shipped," fire an interview with the customers it affected. The team that built it sees the reaction land as an Update on the very item they just closed — closing the build-measure-learn loop without leaving the board.

Support and onboarding milestones. When an onboarding item reaches a milestone status, or a resolved ticket is logged on a board, trigger a short Koji interview to capture the experience while it is fresh. Themes write back so a CS lead can scan sentiment across the whole group at a glance.

What syncs back to monday

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

• Interview status into a Status column (Invited, Completed)
• Structured question scores — a scale satisfaction value, a single_choice reason, a yes/no signal — into Number, Dropdown, and Status columns
• Top themes and a short AI summary posted as an Update (comment)
• A link to the full transcript and Koji report

Because those columns are native monday data, they flow straight into your existing dashboards, filters, and automations — so "customers who rated us ≤ 6" can trigger the next recipe automatically.

Why a conversation beats a status update

Work-management tools are brilliant at tracking whether something got done and terrible at telling you whether it worked for the customer. The feedback loop usually breaks right there: the item is closed, the team moves on, and the customer's actual reaction never makes it back to the board. It matters more than teams admit — Forrester's 2024 US Customer Experience Index found that customer-obsessed companies grow revenue markedly faster than their peers, yet only a small fraction of companies actually qualify as customer-obsessed. The gap is almost always operational: feedback and execution live in different tools.

Koji closes that gap by putting the customer's voice on the same board as the work. Its AI interviewer runs the conversation in voice or text, asks adaptive follow-ups so a rating becomes a reason, and supports all six structured question types in a single interview — open_ended, scale, single_choice, multiple_choice, ranking, and yes_no — so you capture a clean, column-ready signal and the story behind it at once. See the structured questions guide for how to combine them.

Koji's quality gate keeps the board data clean: each session is scored for effort and coherence, so low-effort responses are flagged, excluded, and never consume credits. Koji's free tier includes 10 one-time credits to pilot a loop; paid plans start at €29/mo (Insights) and €79/mo (Interviews), with text interviews costing 1 credit and voice 3.

Common setup mistakes to avoid

Hard-coding the wrong column ID. monday column IDs are per-board — copy the exact ID from the board you are writing to, or the write-back silently lands nowhere.

Interviewing on every status change. Scope the trigger to the specific status that matters (Won, Lost, Shipped) and add a suppression column so the same contact is not invited twice in a short window.

Leaving results as free text. Map Koji's scale and single_choice outputs to Number and Dropdown columns, not one long text column, so your dashboards stay filterable.

Start with one board — monday CRM lost deals or a post-launch project board — prove the write-back, then expand once the loop is landing cleanly.

Related Resources

• Structured Questions Guide — the 6 question types that populate monday columns cleanly
• Linear + Koji Integration — trigger research from engineering workflows
• Jira + Koji Integration — the same pattern for Jira issues
• Notion + Koji Integration — sync research into Notion
• Zapier Research Automation — build the no-code loop
• Make Research Automation — the Make.com alternative
• Webhook Setup — the real-time path
• How the Quality Gate Works — why low-effort responses never pollute your data