- Industry
- Insure AI
- Department
- Marketing
- Team
- Brand and Insights
- Kind
- Investigator
- Approver
- Head of Communications
- Code
- MKT-12
It explains spikes in negative mentions of the insurer, such as a viral claim rejection story, complaints after a catastrophe event or a regulator fine report, by separating the stories driving them from background noise. It logs the diagnosis in the communications tracker and drafts a holding statement when reach passes the crisis threshold. **The problem it solves:** Why negative sentiment or mentions of the insurer changed: the terms, sources and period, the figures against the baseline, when it started and which claims, rate, catastrophe or regulator stories are involved.
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 social listening tool: mention volume, sentiment and top stories by source for the terms and period, with the baseline..
- 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
Weighs competing explanations against the evidence and shows which one holds up.
- 05Asks a person
Hands the work to Head of Communications with its reasoning before anything goes out.
- 06Acts
Once approved, it acts in the communications tracker: it logs the issue with its drivers, severity and owner and attaches the holding statement draft. and records what was done.
What it works with
Your knowledge, your systems, your people.
Knows
The communications playbook: monitored terms, baseline sentiment, crisis levels, approved holding statements and who speaks for the company on claims, regulation and catastrophes.
Reads from
the social listening tool: mention volume, sentiment and top stories by source for the terms and period, with the baseline.
Acts in
the communications tracker: it logs the issue with its drivers, severity and owner and attaches the holding statement draft.
Hands back
An object with change_percent, drivers (a ranked list, each with story, source, reach, evidence and confidence), crisis (true or false), route_to (a list of teams) and holding_statement (text or null).
Asks a person when
Reach passes 250000, a story involves a named claimant, a regulator, or a catastrophe event, or a holding statement is drafted.
Settings
Your limits, set by your managers.
| Setting | Default | What it controls |
|---|---|---|
| Baseline window (days) | 30 | Sentiment and volume are compared with this many days before. |
| Negative spike (%) | 40 | Negative mentions this far above baseline need a diagnosis. |
| Reach that makes it a crisis | 250000 | A story reaching more people than this is treated as a crisis. |
| Sources monitored | news, social, review sites, forums, regulator notices | Comma-separated sources it reads. |
| Draft a holding statement | On | When on, a crisis gets a holding statement draft for the approver. |
Limits are settings, not code: a manager changes them without a developer.
Rules it always follows
- Compare against the last 30 days; a rise of 40 percent in negative mentions needs a diagnosis.
- Rank drivers by evidence: one story with heavy reach outranks scattered complaints, and name the story.
- A story about an individual claim never gets a public reply that discusses the claim; route it to the Claims Manager and complaints team.
- Reach above 250000 is a crisis; when on is true, draft a holding statement for the Head of Communications.
- Read only news, social, review sites, forums, regulator notices.
See the Brand Monitoring Agent agent on your own work.
We’ll run it on a sample of your work and show you what your approver would see.