RESEARCH · ANONYMOUS SURVEY
Timing data shows how long people stay; it cannot say why a conversation felt like a person. So EasyGirl asks them. This study is a single optional question — what made the interaction feel natural? — offered after longer live video conversations.
Deterministic 5% sample
The question is offered to a fixed five percent of sessions that lasted at least thirty seconds — no targeting, no picking by hand.
Multi-select, one question
Respondents tick every signal that applied, or close the prompt and answer nothing. Each anonymous fingerprint counts once.
Anonymous by construction
Only option counts are stored — no free text, no identity attributes, and a daily rotating salt behind each fingerprint.
Status
Insufficient data for publication.
The survey is live but still collecting anonymous responses. Results appear here once the sample is large enough, spans enough distinct days, and has passed human review — never before.
Option rates
The share of valid responses mentioning each natural-interaction signal, ranked from most to least chosen.
Valid response count
How many distinct anonymous responses the rates are based on, after de-duplication by fingerprint.
Collection window
The exact date span the responses came from, printed next to every figure.
Plain disclaimers
That these are voluntary, subjective impressions — never identity checks, never face recognition.
Definitions follow the research methodology.
Natural isn’t a feature flag. It’s a person, answering in real time.
Start a Live Conversation