Shopper Research Synthesis Agent
Turns shopper research and store observation into design recommendations.
- Industry
- Retail
- Department
- Design
- Team
- Product Experience
- Kind
- Investigator
- Approver
- E-commerce Director
- Code
- DSN-02
It synthesises usability tests, store observation, exit interviews, session recordings and survey verbatims into findings about why shoppers struggle, such as abandoning at size selection or missing click-and-collect signage in store. It weighs the competing explanations and recommends design changes with the evidence for each. **The problem it solves:** Why shoppers struggle at the point of the journey or store studied, and which explanation the evidence supports.
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: the funnel drop-off, search and filter use and session metrics for the journey studied..
- 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 design library: it adds the findings and recommendations to the research repository and the design backlog. and records what was done.
What it works with
Your knowledge, your systems, your people.
Knows
The research repository: past studies and findings, the shopper personas and missions, the store observation protocol, and the known issues already on the design backlog.
Reads from
the e-commerce platform's analytics: the funnel drop-off, search and filter use and session metrics for the journey studied.
Acts in
the e-commerce platform's design library: it adds the findings and recommendations to the research repository and the design backlog.
Hands back
An object with findings (a list of finding, severity, participants, evidence, data_check agrees, disagrees or none, likely_cause and alternative_cause) and recommendations (a list of finding, change and expected_effect).
Asks a person when
A recommendation changes checkout, pricing display or store layout across the estate, or the research used footage of identifiable shoppers.
Settings
Your limits, set by your managers.
| Setting | Default | What it controls |
|---|---|---|
| Participants needed for a finding | 3 | A finding needs evidence from at least this many shoppers. |
| Severity scale | four_point | The scale findings are rated on. |
| Check findings against analytics | On | When on, each finding is checked against funnel or store data where it exists. |
Limits are settings, not code: a manager changes them without a developer.
Rules it always follows
- A finding needs evidence from 3 or more shoppers; fewer is reported as a signal, not a finding.
- When on is true, check each finding against funnel drop-off or store data and say whether the data agrees.
- Weigh at least two explanations for each problem, such as unclear sizing versus missing stock, and say what separates them.
- Rate severity on the four_point scale by how many shoppers it blocks and whether it costs a sale.
- Quote shoppers without names or faces, and never use store observation footage that shows identifiable shoppers.
See the Shopper Research Synthesis Agent agent on your own work.
We’ll run it on a sample of your work and show you what your approver would see.