Campaign Analytics Agent
Reports ROAS and acquisition cost by campaign.
- Industry
- Insure AI
- Department
- Marketing
- Team
- Demand Generation
- Kind
- Investigator
- Approver
- Marketing Lead
- Code
- MKT-09
It explains why cost per bound policy, quote-to-bind conversion or return on ad spend moved by campaign and line of business, separating media changes from pricing changes in the rating engine and tracking faults. It publishes the diagnosis to the BI dashboard and flags when a rating change, not marketing, is the likely cause. **The problem it solves:** Why cost per bound policy, quote-to-bind or ROAS moved: the campaigns and lines, the figures for both periods, when the change began and what changed in spend, targeting, creative, pricing or tracking.
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 analytics warehouse: spend, quotes, binds, premium and attributed touches by campaign and line of business for both periods, with the rating engine change log..
- 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 Marketing Lead with its reasoning before anything goes out.
- 06Acts
Once approved, it acts in the BI dashboard: it publishes the diagnosis note on the campaign performance dashboard. and records what was done.
What it works with
Your knowledge, your systems, your people.
Knows
The marketing measurement playbook: attribution model, targets by line of business, rating change calendar and tracking setup.
Reads from
the analytics warehouse: spend, quotes, binds, premium and attributed touches by campaign and line of business for both periods, with the rating engine change log.
Acts in
the BI dashboard: it publishes the diagnosis note on the campaign performance dashboard.
Hands back
An object with metric, change_percent, causes (a ranked list, each with cause, evidence, confidence and how_to_confirm), by_line_of_business (a list of metric changes), tracking_ok (true or false) and recommended_action.
Asks a person when
The diagnosis names a rating or underwriting change as the cause, or a tracking fault means reported figures were wrong.
Settings
Your limits, set by your managers.
| Setting | Default | What it controls |
|---|---|---|
| Attribution model | position-based | How credit for a bound policy is split across touches. |
| Target cost per bound policy (USD) | 120 | The blended target the diagnosis measures against. |
| Variance worth explaining (%) | 15 | A metric moving less than this between periods is noise. |
| Days allowed for quotes to bind | 14 | The latest days of a period are excluded until quotes have had this long to bind. |
| Reporting currency | USD | All money is reported in this currency. |
Limits are settings, not code: a manager changes them without a developer.
Rules it always follows
- Explain only moves of at least 15 percent; credit binds by position-based attribution.
- Exclude the last 14 days of a period from bind-rate comparisons.
- Check the rating engine change calendar first: a rate or underwriting rule change just before the move outranks a media explanation.
- Check tracking before blaming media: a drop in tracked quotes with steady binds in the policy administration system is a tracking fault.
- Report money in USD against 120 per bound policy.
See the Campaign Analytics Agent agent on your own work.
We’ll run it on a sample of your work and show you what your approver would see.