
PostHog shows what users do, not why. Research triggers close that gap — when a PostHog event fires, Usercall invites the user to a 2-5 min AI-moderated interview. Responses are synthesized into themes, not raw transcripts.
Most teams trigger interviews on too many events, too late, or with no user context. That creates noise, burns invitation volume, and gives you vague answers disconnected from the product moment that caused the behavior.
Good posthog user interviews start with high-signal events tied to a real decision, friction point, or drop-off. If an event does not represent a meaningful moment, it should not trigger research.
Add the Usercall script to your product in the same
block as PostHog's JS SDK.This is required for trait-based filters and interview history.
window.usercall.identify({
userId: "u_123",
email: "user@example.com",
traits: { plan: "trial", country: "US" }
});
The allowlist is the exact set of event names that can trigger interviews — keep it narrow.
window.__usercallSetup = window.__usercallSetup || {};
window.__usercallSetup.bindPostHog = {
allowlist: ["onboarding_completed"]
};
Pick the PostHog event, attach an interview guide, and set targeting and frequency limits. Usercall handles delivery — users see an invitation immediately after the event fires.
Use where when the event name alone is too broad. It filters on the source side before events reach Usercall — excluded events don't appear in history or suggestions.
Trait-based rules require identify() to run first.
window.__usercallSetup.bindPostHog = {
allowlist: ["newly_signed_up"],
where: {
properties: { entrypoint: "pricing_page" },
traits: { plan: "trial" }
}
};
This triggers interviews only for trial users who signed up via the pricing page, not every signup.
Use this instead of the PostHog binding if you want to trigger from your own app code.
window.usercall && window.usercall.trigger("pricing_feedback_requested", {
reason: "too_expensive",
page: "pricing"
});
Configured per trigger in the Usercall dashboard. Usercall POSTs the matched event payload, trigger run IDs, and generated interview URL to your endpoint — add a signing secret and requests include an x-usercall-signature header.
This is useful when you want downstream automation in Slack, CRM workflows, or internal tooling the moment a qualified interview invitation is created.
Interviews fire at the moment the event fires, while context is fresh. That is the difference between a user remembering exactly why they quit setup and giving you a generic answer two weeks later.
Usercall turns behavioral events into research-grade qualitative analysis at scale. Instead of sorting through isolated responses, you get recurring themes across users, with AI-moderated interviews, deep researcher controls, and intercepts tied to key product moments where metrics alone stop being useful.
If you are instrumenting PostHog already, this is the cleanest way to attach the “why” to onboarding drop-off, failed activation, cancellation, and feature confusion. For adjacent setups, see PostHog User Feedback, Why Users Drop Off During Onboarding, Churn Interview Questions, and When to Ask Users for Feedback.
Related: PostHog User Feedback · Why Users Drop Off During Onboarding · Churn Interview Questions · When to Ask Users for Feedback
Usercall runs AI-moderated user interviews that collect qualitative insights at scale, with the depth of a real conversation and without the overhead of a research agency. If you want posthog user interviews that trigger from real product behavior and produce usable themes fast, see how Usercall’s AI interview workflows fit into your event pipeline.