Content Performance Agent
Reports engagement and conversion by piece.
- Industry
- Oil & Gas
- Department
- Content
- Team
- Content Planning
- Kind
- Investigator
- Approver
- Head of Corporate Communications
- Code
- CON-12
It reports engagement and conversion by piece for each audience, such as station-locator use and fuel-card applications for customers, report downloads for investors and applications for careers, and explains why pieces over- or under-perform. It separates search, distribution, timing, news-cycle effects such as price spikes or incidents, and tracking causes, recommends a fix and updates the content performance report. **The problem it solves:** Why the pieces named performed as they did: the metrics against their benchmark, when the change began, and what changed in publishing, search, tracking or the news around then (price moves, incidents, results).
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: sessions, engagement and conversion events by piece, channel and audience for the period and the one before..
- 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 Corporate Communications with its reasoning before anything goes out.
- 06Acts
Once approved, it acts in the content reporting dashboard: it updates the content performance report with the period's figures, verdicts and causes. and records what was done.
What it works with
Your knowledge, your systems, your people.
Knows
The performance guide: benchmarks per audience and format, the conversion events, the news calendar (price moves, incidents, results) and known tracking changes.
Reads from
the analytics warehouse: sessions, engagement and conversion events by piece, channel and audience for the period and the one before.
Acts in
the content reporting dashboard: it updates the content performance report with the period's figures, verdicts and causes.
Hands back
A list of pieces, each with piece_id, verdict (under, on or over target, or no verdict), metrics, causes (ranked, each with cause, evidence and confidence) and fix.
Asks a person when
The report attributes a change to an incident, a partner or a market event, or recommends pulling or changing a published piece.
Settings
Your limits, set by your managers.
| Setting | Default | What it controls |
|---|---|---|
| Under-performing below (% of benchmark) | 70 | A piece below this share of its benchmark is under-performing. |
| Out-performing above (% of benchmark) | 130 | A piece above this is out-performing. |
| Fewest sessions for a verdict | 500 | Below this no verdict is given. |
| Conversion events | station_locator_use, fuel_card_application, report_download, careers_apply | Comma-separated. |
| Site-wide drop that points to tracking (%) | 30 | A drop this size across all pieces at once is checked as a tracking fault first. |
Limits are settings, not code: a manager changes them without a developer.
Rules it always follows
- Give no verdict below 500 sessions.
- Mark a piece under-performing below 70 percent and out-performing above 130 percent of its benchmark.
- Check a drop above 30 percent across all pieces as a tracking fault before any content cause.
- Separate news-driven traffic (price spikes, incidents, results) from the piece's own performance before crediting the content.
- Count only station_locator_use, fuel_card_application, report_download, careers_apply as conversions.
See the Content Performance Agent agent on your own work.
We’ll run it on a sample of your work and show you what your approver would see.