toolathon123/south-america-ecommerce-ux-monthly

South America E-Commerce UX Monthly Dataset — v2026-08

Longitudinal monthly dataset of e-commerce user experience (UX) signals collected from major retail platforms in Brazil and Argentina. This is the August 2026 monthly release, intended for academic reproducibility, cross-month comparison, and the publication of monthly UX insight reports by the São Paulo (USP) & Buenos Aires (UBA) collaborative research team.

Quick Access

• Version tag: v2026-08
• Data file: ecommerce_ux_monthly_2026-08.csv (UTF-8 encoded)
• License: CC BY-NC 4.0
• Sample size: 520 records (295 Brazil · 225 Argentina)

1. Data Collection

Attribute	Value
Region	South America — Brazil 🇧🇷 and Argentina 🇦🇷
Languages	Portuguese (Brazil) and Spanish (Argentina)
Time range	2026-08-01 to 2026-08-31 (local time, UTC−03:00 / UTC−04:00)
Sample size	520 records (this monthly release)
Platforms	Mercado Livre, Magalu, Amazon.com.br, Shopee, Americanas, Casas Bahia (BR); Mercado Libre, Falabella, Frávega, Garbarino, Tiendanube, Sodimac (AR)
Device types	desktop, mobile, tablet
Collection method	Voluntary in-app/on-site UX feedback widget + anonymized session interaction logs
Ethics approval	CEPSH–USP No. 2026-0147 (Brazil) · CEI–UBA No. 2026-089 (Argentina)

2. Field Descriptions

Field	Type	Description
user_id	string	Anonymous user identifier (e.g. UX-822337-RB). No PII is recoverable.
country	string	Brazil or Argentina.
city	string	City-level location only (never street/neighborhood).
platform	string	E-commerce platform where the session occurred.
session_timestamp	string	ISO-8601 session start time, local timezone offset.
device_type	string	desktop, mobile, or tablet.
page_dwell_time_s	int	Time spent on the page in seconds.
click_heatmap_zone	string	UI region of the main interaction (e.g. header_search, checkout_cta).
task_completion_rate_pct	int	Task completion rate in percent (0–100).
nps_score	int	Net Promoter Score on a 0–10 scale.
feedback	string	Open-ended feedback in Portuguese or Spanish (UTF-8).

3. Anonymization & Privacy

All personal data has been removed before publication:

• Names, e-mail addresses, and street-level addresses are stripped.
• Users are referenced only by a non-reversible anonymous user_id.
• Location is retained at city level only.
• Free-text feedback was scanned with a PII filter (regex + local entity recognizer) and any residual personal identifiers were replaced with [REDACTED] before release.

No authentication, account, or payment credentials are included.

4. Suggested Uses

• Monthly UX benchmarking and trend analysis across countries/platforms.
• NPS and task-completion modeling (e.g., regression, classification).
• Multilingual (pt/es) customer feedback mining and sentiment analysis.
• Reproducibility of the collaborative team's monthly UX insight reports.

5. Versioning & Citation

Each monthly release is tagged (e.g., v2026-08) for traceability and citation.

@misc{sa_ecommerce_ux_monthly_2026_08,
  title        = {South America E-Commerce UX Monthly Dataset -- August 2026 (v2026-08)},
  author       = {{USP--UBA E-Commerce UX Research Group}},
  year         = {2026},
  month        = {8},
  howpublished = {Hugging Face Datasets: toolathon123/south-america-ecommerce-ux-monthly},
  note         = {CEPSH--USP No. 2026-0147; CEI--UBA No. 2026-089}
}

6. License

This dataset is released under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license, permitting academic sharing and reuse with attribution while restricting commercial use. See https://creativecommons.org/licenses/by-nc/4.0/.