Campaign Landing Page Agent
Tests promotion and campaign landing pages to lift conversion.
- Industry
- Retail
- Department
- Marketing
- Team
- Demand Generation
- Kind
- Decision-maker
- Approver
- E-commerce Director
- Code
- MKT-06
It designs A/B tests for promotion and campaign landing pages on the online store (hero, product grid order, offer banner, filters shown) and reads the results on conversion rate and revenue per visitor, with significance and sample-size checks. It decides whether to start, extend, stop or ship each test and promotes the winning variant in the experimentation tool, never running a test that shows shoppers different prices.
How it works
7 steps, and a person signs off before step 7.
- 01Looks it up
Finds the record the request is about in the experimentation tool: visitors, orders, revenue and guardrail metrics by variant for the test named..
- 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.
- 04Forecasts, when it decides
Projects what each option will cost or return before it recommends one.
- 05Works it through
Prepares the draft, match, recommendation or reply, with its sources.
- 06Asks a person
Hands the work to E-commerce Director with its reasoning before anything goes out.
- 07Acts
Once approved, it acts in the experimentation tool: it sets the traffic split or promotes the winning variant as approved. and records what was done.
What it works with
Your knowledge, your systems, your people.
Knows
The landing page testing playbook: the test design standard, primary and guardrail metrics for retail pages, the promotional calendar's no-test windows and the rule that every variant shows the same price and offer.
Reads from
the experimentation tool: visitors, orders, revenue and guardrail metrics by variant for the test named.
Acts in
the experimentation tool: it sets the traffic split or promotes the winning variant as approved.
Hands back
An object with call (start, extend, stop or ship), variants (a list, each with name, visitors, conversion_rate, revenue_per_visitor and lift_percent), confidence_percent, guardrails (a list, each with metric and change) and reason.
Asks a person when
A winner is to be shipped on a page carrying a live promotion, a test is stopped early, or a guardrail metric worsened.
Settings
Your limits, set by your managers.
| Setting | Default | What it controls |
|---|---|---|
| Confidence needed (%) | 95 | A result counts only at or above this confidence. |
| Least visitors per variant | 5000 | A test is not read before each variant has this many visitors. |
| Shortest test (days) | 7 | A test runs at least a full week so weekday and weekend shopping are both covered. |
| Least lift to ship (%) | 2 | A winner must lift the goal metric at least this much. |
| Do not start tests during | Black Friday to Cyber Monday, Boxing Day sale week | Peak trading days when no new test starts on a landing page. |
Limits are settings, not code: a manager changes them without a developer.
Rules it always follows
- Every variant shows the same prices, offers and stock; a test never varies price.
- Read a test only after 5000 visitors per variant and 7 days, at 95 percent confidence.
- Ship a variant only if it lifts the goal by at least 2 percent without hurting basket value or returns rate.
- Never start a test during Black Friday to Cyber Monday, Boxing Day sale week.
- Stopping a test early or shipping a winner on a page that carries a live promotion goes to the E-commerce Director.
See the Campaign Landing Page Agent agent on your own work.
We’ll run it on a sample of your work and show you what your approver would see.