For information about how REST API works, see rest main page.
Read-only, per-product aggregated counters computed from product interactions. This is how you find out which products are actually performing, without pulling the raw event log.
The same data is exposed over six rolling time windows, each at its own URL:
app.personyze.com/rest/products_stats_1day
app.personyze.com/rest/products_stats_2day
app.personyze.com/rest/products_stats_4day
app.personyze.com/rest/products_stats_week
app.personyze.com/rest/products_stats_recent
app.personyze.com/rest/products_stats_allAll six have identical columns. GET is the only supported method — the counters are maintained by Personyze's summary pipeline and are not writable.
Columns
Rows are keyed by internal_id, your own product identifier.
Raw counts in the window, one per interaction status:
n_viewed, n_delivered, n_extra, n_favorite, n_goal
Account-wide comparison figures, for judging one product against the rest of the catalog. For each status X in the list above:
| Column | Meaning |
|---|---|
overall_n_X |
Total count across the account. |
overall_avg_X |
Average per product, counting products with zero. |
overall_avg_nonzero_X |
Average per product, counting only products with at least one. |
percentile_X |
Where this product falls against the rest — the useful column for ranking. |
Recent conversions: last_goal_user_id_0..2 and last_goal_user_time_0..2 hold the three most recent users to reach goal on this product, with their timestamps.
Select example
The 20 best-converting products of the last week:
curl 'https://api:FD09908F25BECB09D78EBEA1DADFB618651F612D@app.personyze.com/rest/products_stats_week/order_by_desc/n_goal/limit/20'One product's all-time numbers:
curl 'https://api:FD09908F25BECB09D78EBEA1DADFB618651F612D@app.personyze.com/rest/products_stats_all/where/internal_id=SKU-8891'Just the columns you need:
curl 'https://api:FD09908F25BECB09D78EBEA1DADFB618651F612D@app.personyze.com/rest/products_stats_recent/columns/internal_id,n_viewed,n_goal,percentile_goal/order_by_desc/percentile_goal/limit/50'Freshness
These are batch-computed aggregates, not live counters. A purchase posted a minute ago will appear in the raw product interactions immediately, but in these windows only after the next summary run.