When Measures Become Targets

wèile为了pànduàn判断yī一xiàng项fúwù服务shìfǒu是否gǎishàn改善，guǎnlǐ管理zhě者tōngcháng通常xūyào需要kě可bǐjiào比较de的zhǐbiāo指标。

To assess whether a service is improving, managers usually need comparable measures.

děngdài等待shíjiān时间、chǔlǐ处理shùliàng数量hé和mǎnyì满意dù度，dōu都néng能bǎ把móhu模糊de的yìnxiàng印象zhuǎnhuà转化wèi为jiào较wèi为míngquè明确de的xìnxī信息。

Waiting times, case volumes and satisfaction ratings can turn vague impressions into more concrete information.

dàn但dāng当zhǐbiāo指标zhíjiē直接juédìng决定jiǎnglì奖励shí时，cānyù参与zhě者yě也kěnéng可能wéirào围绕zhǐbiāo指标tiáozhěng调整xíngwéi行为。

But when metrics directly determine rewards, participants may adapt their behaviour around them.

lìrú例如，zhǐ只qiángdiào强调chǔlǐ处理sùdù速度，kěnéng可能cùshǐ促使gōngzuò工作rényuán人员yōuxiān优先wánchéng完成jiǎndān简单shìxiàng事项。

For example, focusing only on processing speed may encourage staff to prioritise simple cases.

fùzá复杂xūqiú需求yīnér因而bèi被yán延hòu后，jíshǐ即使píngjūn平均děngdài等待shíjiān时间kànqilai看起来yǒusuǒ有所xiàjiàng下降。

Complex needs may then be delayed, even if the average waiting time appears to fall.

wèntí问题bìng并bù不shì是shùzì数字tiānrán天然bù不kěxìn可信，érshì而是shùzì数字zhǐnéng只能miáoshù描述mùbiāo目标de的yībùfen一部分。

The problem is not that numbers are inherently unreliable, but that they describe only part of the goal.

yīdàn一旦zhè这bùfen部分bèi被dàngzuò当作quánbù全部，wèi未bèi被cèliáng测量de的jiàzhí价值biàn便róngyì容易bèi被hūlüè忽略。

Once that part is treated as the whole, unmeasured values can easily be neglected.

gǎishàn改善píngjià评价tǐxì体系，shǒuxiān首先xūyào需要míngquè明确zhǐbiāo指标yuánběn原本fúwù服务yú于shénme什么mùdì目的。

Improving an evaluation system first requires clarifying the purpose the metrics were intended to serve.

qícì其次，yīngdāng应当guānchá观察bùtóng不同zhǐbiāo指标zhījiān之间shìfǒu是否cúnzài存在chōngtū冲突，ér而bù不shì是yāoqiú要求měi每gè个shùzì数字tóngshí同时shàngshēng上升。

Next, it is necessary to examine conflicts between measures rather than demand that every number rise at once.

dìng定liàng量jìlù记录kěyǐ可以yǔ与gèàn个案fēnxī分析jiéhé结合，yǐ以jiěshì解释yìcháng异常jiēguǒ结果hé和yǐncáng隐藏de的dàijià代价。

Quantitative records can be combined with case analysis to explain unusual results and hidden costs.

yīxiàn一线rényuán人员de的fǎnkuì反馈yě也zhíde值得bǎoliú保留，yīnwei因为tāmen他们wǎngwǎng往往zuì最zǎo早chájué察觉guīzé规则yǔ与shíjì实际xūqiú需求de的cuòwèi错位。

Front-line feedback is also worth retaining, as staff often notice mismatches between rules and actual needs first.

bùguò不过，jiārù加入gèng更duō多zhǐbiāo指标bìng并bùbì不必rán然jiějué解决wèntí问题，guòdù过度jìlù记录běnshēn本身yě也huì会zhànyòng占用fúwù服务shíjiān时间。

However, adding more metrics does not necessarily solve the problem; excessive record-keeping itself takes time away from service.

hélǐ合理de的píngjià评价yīngdāng应当yǔnxǔ允许xiūzhèng修正，bìng并chéngrèn承认bùfen部分zhòngyào重要jiàzhí价值nányǐ难以jí即shí时liàng量huà化。

Sound evaluation should allow revision and acknowledge that some important values are difficult to quantify immediately.

zhǐbiāo指标zuì最yǒuyòng有用de的shíhou时候，shì是tā它bāngzhù帮助rénmen人们tíchū提出gèng更hǎo好de的wèntí问题，ér而bù不shì是tì替rénmen人们jiéshù结束pànduàn判断。

Metrics are most useful when they help people ask better questions, rather than bring judgement to an end.