Content Performance Agent
Reports engagement and conversion by piece.
- Industry
- Retail
- Department
- Content
- Team
- Content Planning
- Kind
- Investigator
- Approver
- E-commerce Director
- Code
- CON-12
It reports engagement, add-to-basket and attributed sales for each content piece and explains why a piece under- or over-performed, separating content quality from stock-outs, price changes, traffic source and promotion timing. It recommends what to fix, refresh or stop. **The problem it solves:** Why each content piece performed above or below its benchmark in the period, given stock, price, traffic and promotion effects.
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 e-commerce platform's analytics: sessions, add-to-basket, content-assisted orders and traffic source by piece, with stock and price changes on the featured SKUs..
- 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 E-commerce Director with its reasoning before anything goes out.
- 06Acts
Once approved, it acts in the e-commerce platform's content calendar: it updates each piece with its performance verdict and the recommended action. and records what was done.
What it works with
Your knowledge, your systems, your people.
Knows
The content measurement model: attribution rules for content-assisted sales, the benchmarks by format and category, and the known distortions such as stock-outs, markdowns and sale-event traffic.
Reads from
the e-commerce platform's analytics: sessions, add-to-basket, content-assisted orders and traffic source by piece, with stock and price changes on the featured SKUs.
Acts in
the e-commerce platform's content calendar: it updates each piece with its performance verdict and the recommended action.
Hands back
An object with pieces (a list of item_id, sessions, add_to_basket_rate, assisted_sales, vs_benchmark_percent and verdict over, on, under or too_early), causes (a list of item_id, likely_cause, evidence and confidence) and actions (a list of item_id and fix, refresh or stop).
Asks a person when
The report recommends retiring a piece that earns organic search traffic, or attributes a fall to a pricing or stock problem that belongs to merchandising.
Settings
Your limits, set by your managers.
| Setting | Default | What it controls |
|---|---|---|
| Attribution window (days) | 7 | A sale counts as content-assisted if the shopper viewed the piece within this many days before buying. |
| Below benchmark that counts as underperforming (%) | 30 | A piece this far below its format and category benchmark is underperforming. |
| Sessions needed before judging a piece | 1000 | A piece with fewer sessions is reported as too early to judge. |
Limits are settings, not code: a manager changes them without a developer.
Rules it always follows
- Credit a sale to a piece only when the shopper viewed it within 7 days before buying.
- Judge a piece against its format and category benchmark; below by 30 percent or more is underperforming.
- Do not judge a piece with fewer than 1000 sessions.
- Before blaming the content, check whether its featured SKUs were out of stock, re-priced or marked down in the period, and whether a sale event skewed traffic.
- Weigh at least two explanations and say what evidence separates them.
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.