Policyholder Research Agent
Plans research with policyholders, claimants and brokers and synthesises what they say.
- Industry
- Insure AI
- Department
- Product
- Team
- Discovery and Delivery
- Kind
- Assistant
- Approver
- Compliance Officer
- Code
- PRD-07
It plans research studies with policyholders, claimants, agents and brokers (method, screener, sample, discussion guide and incentive) and synthesises interview notes into themes backed by participant counts and quotes, such as confusion over excess or delays after first notice of loss. It files the study plan or synthesis in the document system with consent records and keeps health and claim details out of the shared output.
How it works
6 steps, and a person signs off before step 6.
- 01Looks it up
Finds the record the request is about in the policy administration system: anonymised panel candidates by product, tenure and channel, with an open-claim flag and their consent to be contacted for research..
- 02Reads the request
Works out what is being asked and what a complete answer needs.
- 03Checks the facts
Checks figures, dates and rules against the source records, not from memory.
- 04Works it through
Prepares the draft, match, recommendation or reply, with its sources.
- 05Asks a person
Hands the work to Compliance Officer with its reasoning before anything goes out.
- 06Acts
Once approved, it acts in the document and correspondence system: it files the study plan or synthesis with its consent log and retention date. and records what was done.
What it works with
Your knowledge, your systems, your people.
Knows
The research ops guide: consent wording, incentive rules, the recruitment panel, rules for researching vulnerable customers and claimants, data protection for health data and past study findings.
Reads from
the policy administration system: anonymised panel candidates by product, tenure and channel, with an open-claim flag and their consent to be contacted for research.
Acts in
the document and correspondence system: it files the study plan or synthesis with its consent log and retention date.
Hands back
An object with stage, plan (method, groups, sample_size, screener, discussion_guide, incentive_total_usd and consent_notes, or null) and synthesis (a list of themes, each with theme, participants_by_group, quotes and implication, or null), plus open_questions (a list).
Asks a person when
The study's incentives exceed 1500 USD, it involves claimants or customers who may be vulnerable, or it would collect health data.
Settings
Your limits, set by your managers.
| Setting | Default | What it controls |
|---|---|---|
| Participants per group | 6 | Each participant group is recruited to at least this many people. |
| Incentive per participant (USD) | 50 | The standard thank-you per session; brokers and agents are not paid an incentive. |
| Most incentive spend per study (USD) | 1500 | A study plan costing more than this needs the approver. |
| Do not recruit claimants with an open claim | On | When on, claimants whose claim is still open are not recruited, to avoid influencing the claim. |
| Participants for a theme | 3 | A theme needs at least this many participants behind it. |
| Keep raw notes and recordings for (days) | 90 | Raw material is deleted after this many days; the synthesis is kept. |
Limits are settings, not code: a manager changes them without a developer.
Rules it always follows
- Recruit at least 6 per group and pay 50 USD per session; an incentive is never tied to the participant's policy, premium or claim outcome.
- When on is true, never recruit a claimant whose claim is still open.
- Take recorded consent before every session; a participant who may be vulnerable is offered a supported session and may stop at any time.
- Never put health data, claim amounts or policy numbers in the synthesis; refer to participants only by id and group.
- A theme needs at least 3 participants and at least one quote; say how many in each group hold it.
- Delete raw notes and recordings after 90 days.
See the Policyholder Research Agent agent on your own work.
We’ll run it on a sample of your work and show you what your approver would see.