get_conversion_correlations
Statistically vetted factors that move conversion up or down, using two-proportion z-tests with Benjamini-Hochberg correction across all factors tested.
When to use it
The rigorous answer to "why aren't visitors converting". Prefer it over hand-comparing segments — and treat it as the arbiter when it disagrees with your eyeballing.
Ask for it like this
You do not name the tool. These are questions that make an assistant reach for it:
What surprising factor predicts whether someone buys?
Do JavaScript errors actually cost me conversions?
How to read the result
- Only findings that survive multiple-comparison correction are returned, so a result here is not an artefact of scanning many factors.
- The window adapts (7, 14 or 30 days) to reach enough data, so it may not match the window you asked for.
- Factors covered: device type, UTM source and medium, entry page type, JS errors present, pages viewed, engagement level, frustration level, and traffic source.
Parameters
factorName
Restrict to one factor: device_type, utm_source, utm_medium, entry_page_type, had_js_errors, pages_viewed, engagement_level, frustration_level or traffic_source.
confidence
strong is p<0.01, moderate p<0.05, weak p<0.10.
strongmoderateweaklimit
Maximum correlations to return.
Shared parameters it accepts
Documented in full on the tool reference index.
Date window — defaults to the last 30 days.
dateFrom
Start of the window, YYYY-MM-DD. Inclusive.
dateTo
End of the window, YYYY-MM-DD. Inclusive.
Every tool also accepts merchantId, which resolves from your connection. You never pass it.
What comes back
JSON. These are the top-level keys; date-scoped tools additionally carry a _meta block with the server date and the data coverage envelope.
correlations
Each describes a session segment, its conversion rate, the baseline, the lift or drop, and the confidence grade.