NOTES · PRIORITIES
Ask group make dem list wetin dey wrong and den vote on wetin matter pass, and di winner na almost always di tin wey evribodi fit see. Warehouse go name picking errors. Emergency department go name waiting times. Ministry go name public trust. Each of dem dey real, and each one na result: sometin wey dey further back dey produce am.
Kevin Dye give dis one name: di erroneous priorities effect. For complex situations, di problems wey pipo rank pass na always di ones wey dey furthest downstream. Wetin dey pain pass na wetin dey show pass, and wetin dey show pass no dey usually be wia e start. Di one wey dey shine dey opposite to di one wey deep.
John Warfield add di second half, spreadthink: put fifteen pipo wey sabi di matter for one room and dia own priorities no go too overlap. No be say anybodi wrong, but becos complex situation dey hold more than wetin one pesin fit hold. For our runs, 61–89% of all di items wey dem propose dey get at least one vote. To scatter na di normal state of room, e no be failure.
You no fit score real room, becos no real room get answer key. Simulation fit. We build one organisation wey get twenty problems, and five of dem na really di deep causes, and we make fifteen pipo work through am two hundred times for each setting, dey decide each pair di way True Causes dey do am: one vote each, wey dey settle at 75% of di votes wey dem cast.
Wen wetin dey pain no mislead di room, to stop for di vote dey name 3.1 out of di 5 real causes, while to build map of wetin dey affect wetin dey name 3.9. Even room wey no get bias dey do betta with map, becos pair question at 75% na betta instrument than ranking.
Wen di room dey rank only by wetin e feel, di vote dey name none of di five, and to build map on top dey rescue only 0.2. Cause wey dem drop for di ballot no fit come back later: once item fail di vote, no amount of careful arranging go bring am back.
Voting dey fine to choose among options. Na di wrong tool to find causes, becos e dey measure wetin dey shine and e dey call am priority. Di fix no be betta vote. Na to ask different question: for each pair of problems, one get big hand for di oda one? Di deepest causes na di ones wey evritin else dey trace back to, and dem be often di cards wey nobodi vote for.