Skip to main content

Introduction

Cafe24 is a Korean e-commerce platform that helps merchants run their direct-to-consumer storefronts. Cafe24 Analytics is the analytics surface of that platform — the pre-aggregated session, sales, product, ad-effect, and visit-path numbers that mall owners see inside Cafe24’s own analytics dashboard. This connector reads from the Cafe24 Data API (cafe24data.com), not the commerce Admin REST API. The data it returns is the same set of pre-aggregated reports Cafe24 ships in the merchant console: visitor sessions, page views, traffic sources, daily active visitors, product views and sales, cart actions, payment methods, sales-per-visitor analyses, and visit-path attribution. Many teams pair it with the Cafe24 commerce connector, which surfaces raw orders and products. As an analytics connector, session, event, and breakdown data are standardized so widgets can aggregate alongside ad-platform sources without additional configuration. To connect this data source, see How to connect Cafe24 Analytics to Adriel.

Data refresh strategy

Architecture data

Mall discovery and shop metadata are fetched at connection time and refreshed alongside the OAuth token. The Cafe24 Analytics endpoint catalog itself is registered statically in the connector (23 report endpoints plus six computed metrics) and does not change without a connector update.

Reports data

Cafe24 Analytics is a daily-cache connector backed by a JSON daily cache:
  • Initial fetch: when the data source is created, the last 90 days of daily data are loaded into the cache.
  • Daily refresh: the cache is refreshed twice daily, at 02:00 UTC and 14:00 UTC. Each cycle re-syncs the most recent 3 days of data.
  • History window: data older than the 90-day initial backfill is not available from cache.
Worker concurrency is forced to one at both job and group levels, and outgoing requests are rate-limited to one request per 500 ms per asset to stay within Cafe24’s quotas. Future-dated rows are clamped at fetch time.

Architecture levels

Mall (mall_id) → Shop / User (user_id) → Report endpoint → Row Each Cafe24 Data API report endpoint is a separate reporting surface; a single query targets one endpoint and returns its rows for the selected date range.

Date range limits

These caps are enforced per query and depend on the selected time breakdown. Longer historical pulls must be split into multiple queries, and data older than the 90-day initial backfill is not retained in cache.

Metrics

How to read the columnsCode style like visitors_unique__unique_visit_count is the connector’s field key, following the endpoint__field pattern where the endpoint is the report family and the field is the column. Italic text marks a value computed by Adriel (a derived metric) rather than returned directly by the Cafe24 Data API.Data type vocabulary: Number, Currency, Percentage, Ratio, Duration, Date, Text, URL, Array, Boolean. Currency values are reported in the mall’s settlement currency (KRW for most Cafe24 malls).

Users & sessions

Engagement

Traffic acquisition

Events & conversions

Ecommerce

Pages & screens

Breakdowns

How to read the columnsEach Cafe24 Analytics report endpoint exposes exactly one grouping column, so a widget selects one report breakdown at a time. Code style like products_sales is the connector’s endpoint breakdown key. Italic text marks a breakdown produced by Adriel rather than by the Cafe24 Data API.

Visitors

Pages

Products

Cart

Sales

Visit paths

Ad effect

Time

Adriel groups Cafe24 Analytics’ daily-granularity results into these standard views for reporting alongside other connectors.

General

These grouping options are provided by Adriel and apply across connectors.

Limitations

  • One breakdown per query. Each endpoint exposes exactly one grouping column. Widgets that combine more than one non-time breakdown are rejected with INVALID_DIMENSIONS_METRICS_COMBINATION. The reserved channel value is stripped internally before execution.
  • Metrics depend on the selected breakdown. Because each report family is a separate endpoint, the available metrics change with the chosen breakdown; metrics from one endpoint cannot be mixed with a different endpoint’s breakdown.
  • 90-day history. Only the last 90 days are backfilled into cache at connection time; older data is not available.
  • Per-query date-range caps. Daily queries cap at 60 days, weekly at 12 weeks, and monthly at 12 months. Longer ranges must be split.
  • Throttled, serialized fetching. Requests are paced to one per 500 ms per asset and worker concurrency is forced to one, so large queries are paced rather than parallelized.
  • Pre-aggregated source. Values are Cafe24’s own pre-aggregated dashboard numbers and may differ from raw order-level figures surfaced by the Cafe24 commerce connector.

API references

See also