You'll triage a deck of 50 emails twice — once as a swipe deck, once as a click-through list — in a randomized, counterbalanced order. Each email shows the action to take; just execute it as fast and accurately as you can. We record per-email time and misfires, then show you the result and a CSV you can keep.
It takes about 4–6 minutes. Best on the device you actually triage email on. This is the exact instrument behind the Flick triage-speed benchmark.
A within-subjects, counterbalanced design removes per-person speed as a confound and cancels learning effects. Each email carries a target disposition so this measures the interaction cost (locate → aim → act vs. one gesture), holding the decision constant — exactly the question the benchmark isolates. It is not a reading-comprehension test.
Condition0 / 50
Your result
Median seconds / email — Click
Median seconds / email — Swipe
Swipe vs. click
Total time to clear 50 — Click
Total time to clear 50 — Swipe
Misfires — Click / Swipe
One run is an anecdote, not a study. The published benchmark needs ~8 people. Copy your CSV and send it to whoever's collecting — pooled medians across participants are the real result.
Caveats this single run can't escape: a list inbox's multi-select (archive 30 newsletters at once) can erase the swipe edge on bulk-heavy inboxes — this benchmark triages one-at-a-time on both sides by design. Keyboard power-users (press e to archive) are a separate, faster class. Report misfires honestly: a fast gesture that archives the wrong email isn't faster.