Modeling the Full Sample Pipeline, Not Just the Screener

The Screener Funnel Simulator handles qualification logic. This tool handles everything else: the operational losses that happen before and after the screener.

An email invitation doesn't reach 100% of the people you send it to. Some bounce. Some land in spam. Of the ones that arrive, maybe 15-25% open the email. Of those who open, maybe 40-60% click through. Of those who click, some abandon before the screener even loads. Then the screener attrition happens. Then there's mid-survey dropout. Then data quality checks remove speeders and straightliners.

By the time you account for every stage, you might need to send 50,000 invitations to get 500 completed surveys. That's a 1% end-to-end yield. And if you only budgeted for 20,000 sends based on a napkin estimate, you're going to run out of sample halfway through fieldwork.

How it differs from the screener tool

The Screener Funnel Simulator focuses on qualification criteria. This tool models the full operational pipeline, including stages that have nothing to do with respondent qualification: email deliverability, open rates, click rates, page-load abandonment, survey fatigue dropout, and post-field data quality exclusions.

Each stage has industry benchmark default values that you can adjust based on your own historical data. If you know your panel's average open rate is 22%, plug that in. If you know your typical mid-survey dropout is 18% for a 20-minute survey, use that number.

Tech stack

React 18.2 + Babel CDN. The pipeline model applies sequential attrition rates with industry benchmark defaults sourced from published panel performance data and my own operational experience. The bottleneck identification highlights which stage has the largest absolute respondent loss.