A churn study asks former customers what happened before they left, what they tried, and why an alternative became more attractive. An AI agent can coordinate that work, but the audience must be people who actually used your product.
This walkthrough uses an authorized customer email list and the current BYOP study tools. A general-population panel is not a substitute for your churned customers.
The workflow at a glance
Choose BYOP → create a draft → customize the churn brief → review the saved plan → approve and send invitations → monitor interviews → retrieve and verify the report.
Step-by-step: the actual tool calls
Step 1: Create the BYOP draft
The user must explicitly choose BYOP before the call:
create_study({"name":"Recent customer churn","recruiting_method":"byop"})Step 2: Customize the research plan
customize_study({"study_id":"<study-id>","message":"Interview customers who cancelled in the last 90 days. Explore their original goal, the sequence leading to cancellation, alternatives they considered, and what might have changed the decision."})Relay the backend’s planning questions to the user. Do not infer cancellation reasons from the CRM. Once the plan is saved, call get_study, show the full plan, and obtain approval.
Step 3: Invite the authorized customers
For a BYOP study, use create_participants with 1–100 unique participant emails per batch after the saved plan and invitations are approved. Invitations send by default; set silent: true on individual participant records when invitations should not send. Source customer lists through your own authorized export or integration; MCP has no direct CRM segment-sync tools.
create_participants({"study_id":"<study-id>","participants":[{"email":"participant@example.com"}]})Replace the example with the authorized customer list. Keep cohort selection and any CRM export in your own approved workflow. A successful participant-creation call does not mean the interview has completed.
Step 4: Monitor interviews
list_interviews({"study_id":"<study-id>","status":"completed","page":1,"page_size":20})Paginate and inspect quality labels before presenting a completion count. Report the actual number interviewed and the invitation cohort separately. Participation can be selective; people who respond may differ from those who do not.
Step 5: Analyze and retrieve the study report
generate_report({"study_id":"<study-id>"})
get_study_report({"study_id":"<study-id>"})Study results expose findings, participant responses, sample profiles, recommendations, and source references in JSON. Use generate_report when analysis is needed, and get_interview to verify supporting messages and recording links. Preference shares, credibility scores, and ranked themes are not guaranteed typed fields in this response.
What do you get out the other end?
A useful churn readout distinguishes the initial problem, triggering event, failed recovery attempts, and final decision. Ask the agent to attach evidence to each finding and show exceptions. A stated price objection may reflect low perceived value or an implementation problem; interview evidence should determine the interpretation.
The report is not a statistical estimate of all churned customers. Explain the cohort, number of interviews, missing perspectives, and any contradictions before recommending a product or retention change.
Following up with earlier studies
Agents can search findings and participant responses across authorized studies, then retrieve the underlying reports and interviews. Results preserve study context and source links; the calling agent interprets the evidence. Search coverage is explicit, and retrieving evidence does not launch research.
For a quarter-to-quarter comparison, select the relevant study IDs, retrieve each report, and inspect interviews behind important claims. Distinguish changes in the audience from changes in the product experience.
Why your own customers matter
Someone recruited for general SaaS experience can discuss category expectations. They cannot explain what happened in a relationship they never had with your company. BYOP keeps the research tied to the actual churn event and makes findings useful for your customer success and product teams.
See agentic research, the MCP reference, and the churn analysis solution.