True Causes
ZH
← 大厅
7/7 · 行动
30 人 · 27 个想法
点击任意已完成的阶段进行查看
✓
提出
✓
澄清
✓
归类
✓
选择
✓
结构
✓
地图
7
行动
演示 · 不计入
障碍 · 持有链接的任何人 · 公开列出 · 已开启联盟
主持人:它的主持人(已封存)
30 人 · 27 个想法 · 2 个小组
通行证 → · 逻辑 → · 记录
↗ 分享这张地图
在其他网站上,作为实时地图:
作为图片: https://truecauses.org/d/b3bc98a/map.svg
作为数据: 本房间每一个已判断的卡片对(CSV). 每个主张都有自己的地址 /c/b3bc98a/<card>,列在地图下方。
发布时请注明出处: Causal map: True Causes, https://truecauses.org/d/b3bc98a
问题

Why do complaints keep rising while the product improves?

提出该做什么,再判断每项行动能否撼动每个根源。

封存于 23 天前:这个房间已不再知道谁曾在其中;只有主持人仍然留名。它得出的结论保留。
30 人在座 (其中 29 名为模拟成员) · 针对 5 个根本原因已提出 3 项行动。添加一项,然后判断每项行动能否撼动每个原因。
正以访客身份查看。输入用户名即可加入并参与。
关闭 ✕
提出的行动
Measure agents on problems solved at first contact, not tickets closed per hour
还没有人发言。
有些主张在一个循环中相互影响
↺ A customer who complains twice gets the same f → Most complaints are about waiting, not about t
循环可能是一项真实的发现——一个自我强化的回路——也可能是一条方向从未真正确定的关联。地图把这组主张视为一个节点,所以其中没有一项单独是根源。如果是后一种情况,值得把这个卡片对重新交给房间。
先给根本原因打分
为什么
本方法自身的做法,在提出任何行动之前:每个根本原因按 1–5 分打分——解决它的可行性有多高、解决它会带来多大影响、不加干预它自行解决的可能性有多大。优先考虑影响大、可行性高、自行解决可能性低的根源——优先级 = 影响 + 可行性 − 可能性。分数在小组内取平均,并显示有多少人打了分。
1 = 低,5 = 高。这是你自己的分数;旁边是小组的平均分。
根本原因可行性影响自行解决优先级
Angry customers are moved to the back of the queue by the priority rules––––
Agents who take the time to fix a problem properly miss their numbers––––
Refund policy is different on the website, in the app and in the script––––
Nobody from product reads the ticket tags––––
Managers see the dashboard, agents see the queue, nobody sees the customer––––
每项行动能撼动每个根本原因吗?
竖列:大家提出的行动。横行:地图找出的根本原因。每个格子都是对一个问题的一次投票——这项行动能否显著减少那个原因?一方必须达到所投票数的 75% 才算定下;达不到,格子就保持分裂。 每个格子显示其投票分布——是 / 更糟 / 否——当其中一项达到四分之三时,该格子就定下了。
为什么
以所投票数的 75% 定下,和这里的其他所有问题一样。分裂表示有人投了票,但没有一方达到这个标准——这是一个结果,不是排队等待,它会一直保持分裂,直到足够多的人改变主意。没有人定下的格子保持未定;它永远不会变成否。逐对单独判断,能防止一项受欢迎的行动被记上它根本没触及的功劳。
行动根本原因 · 左边的行动能减少它吗?
Angry customers are moved to the back of the queue by the priority rulesAgents who take the time to fix a problem properly miss their numbersRefund policy is different on the website, in the app and in the scriptNobody from product reads the ticket tagsManagers see the dashboard, agents see the queue, nobody sees the customer
Measure agents on problems solved at first contact, not tickets closed per hour 当前焦点 ↑ · 关闭 ✕✓ 是,它能减少这一项
24 是 / 0 更糟 / 5 否
尚未确定
尚未确定
✓ 是,它能减少这一项
27 是 / 0 更糟 / 2 否
✓ 是,它能减少这一项
25 是 / 0 更糟 / 4 否
Let agents refund up to a set amount without asking 讨论✗ 不,不能
4 是 / 0 更糟 / 25 否
尚未确定
尚未确定
✗ 不,不能
5 是 / 0 更糟 / 24 否
✗ 不,不能
5 是 / 0 更糟 / 24 否
A weekly hour where product reads the top ten tags with support in the room 讨论✗ 不,不能
4 是 / 0 更糟 / 25 否
尚未确定
尚未确定
✗ 不,不能
5 是 / 0 更糟 / 24 否
✓ 是,它能减少这一项
25 是 / 0 更糟 / 4 否
目前杠杆作用最大
Measure agents on problems solved at first contact, not tickets closed per hour 减少了 5 个根本原因中的 3 个
A weekly hour where product reads the top ten tags with support in the room 减少了 5 个根本原因中的 1 个
地图上的行动
显示的是房间判断过的所有行动:它直接作用于 5 个根本原因中的 3 个。 有 2 个没有直接作用。
只依据房间在行动阶段定下的格子,别无其他。绿色的根本原因至少被所显示行动中的一项减轻;灰色的,没有一项直接作用于它;红色边框表示其中某项会使它更糟。
Angry customers are moved to the back of the queue by the priority rules → Agents who take the time to fix a problem properly miss their numbersA customer who complains twice gets the same first reply both times → Most complaints are about waiting, not about the answerA customer who complains twice gets the same first reply both times → Complaints are counted only when someone uses the word complaintMost complaints are about waiting, not about the answer → A customer who complains twice gets the same first reply both timesMost complaints are about waiting, not about the answer → Complaints are counted only when someone uses the word complaintAgents who take the time to fix a problem properly miss their numbers → Complaints about one feature spike after every releaseAgents who take the time to fix a problem properly miss their numbers → Refund policy is different on the website, in the app and in the scriptManagers see the dashboard, agents see the queue, nobody sees the customer → Survey requests go out before the problem is solvedNobody from product reads the ticket tags → The best agents leave within a yearNobody from product reads the ticket tags → Survey requests go out before the problem is solvedRefund policy is different on the website, in the app and in the script → A customer who complains twice gets the same first reply both timesRefund policy is different on the website, in the app and in the script → Agents cannot issue a refund without a manager, and the manager is in meetingsSurvey requests go out before the problem is solved → The same question is answered a hundred times because the help page is wrongSurvey requests go out before the problem is solved → Templates make every reply sound like nobody read the messageAngry customers are moved to the back of the queue by the priority rules ① Measure agents on problems solved at first contact, not tickets closed per hourAngry customers are movedto the back of the queue bythe priority rules①所显示的行动对这个根本原因的作用1 项行动减轻它Managers see the dashboard, agents see the queue, nobody sees the customer ① Measure agents on problems solved at first contact, not tickets closed per hour ③ A weekly hour where product reads the top ten tags with support in the roomManagers see the dashboard,agents see the queue,nobody sees the customer①③所显示的行动对这个根本原因的作用2 项行动减轻它Nobody from product reads the ticket tags ① Measure agents on problems solved at first contact, not tickets closed per hourNobody from product readsthe ticket tags①所显示的行动对这个根本原因的作用1 项行动减轻它The best agents leave within a yearThe best agents leavewithin a yearSurvey requests go out before the problem is solvedSurvey requests go outbefore the problem issolvedAgents who take the time to fix a problem properly miss their numbersAgents who take the time tofix a problem properly misstheir numbers所显示的行动对这个根本原因的作用没有行动直接作用于它Templates make every reply sound like nobody read the messageTemplates make every replysound like nobody read themessageThe same question is answered a hundred times because the help page is wrongThe same question isanswered a hundred timesbecause the help page iswrongComplaints about one feature spike after every releaseComplaints about onefeature spike after everyreleaseRefund policy is different on the website, in the app and in the scriptRefund policy is differenton the website, in the appand in the script所显示的行动对这个根本原因的作用没有行动直接作用于它Most complaints are about waiting, not about the answerMost complaints are aboutwaiting, not about theanswer属于一个反馈回路⟳A customer who complains twice gets the same first reply both timesA customer who complainstwice gets the same firstreply both times属于一个反馈回路⟳Agents cannot issue a refund without a manager, and the manager is in meetingsAgents cannot issue arefund without a manager,and the manager is inmeetingsComplaints are counted only when someone uses the word complaintComplaints are counted onlywhen someone uses the wordcomplaint
地图上的每个编号标出该行动所减轻的一个根本原因。
① Measure agents on problems solved at first contact, not tickets closed per hour
② Let agents refund up to a set amount without asking
③ A weekly hour where product reads the top ten tags with support in the room
从行动到方案
表格对每项行动是单独判断的。方案是房间会一起执行的一组行动。
成员用能减轻某个根本原因的行动组合出方案,房间再像判断其他一切那样,把方案两两比较,以所投票数的四分之三为准。每个方案都显示它直接作用于哪些根本原因、哪些没有、又使哪些更糟。
主持人可以开启方案比较。
我们预期接下来会发生什么
关于各项行动的预测:每项应改变什么、如何衡量、何时检查。到了那一天,曾在房间里的某个人会在旁边写下实际发生了什么。
为什么
地图说明哪些行动能撼动哪些原因。它不说明是否有人做了这些行动,也不说明是否有什么改变了。这里的每一条都是一项预测,带有日期和要衡量的东西,在地图定下后写成。日期到了,实际发生的情况会写在旁边,而预测保持原样:错误的预测才是有用的那种。没有人对这些投过票,它也不改变房间发现的任何东西。
到期:23 Jan 2027
行动Measure agents on problems solved at first contact, not tickets closed per hour
预期fewer people have to come back a second and third time
衡量指标% of contacts resolved at first contact
估计房间:59(15 人的中位数,范围 45 到 72)