社會(huì)的飛速發(fā)展給許多行業(yè)帶來了新的機(jī)遇，這些行業(yè)越來越趨向依靠大數(shù)據(jù)的分析做出決策。那么如何利用分析學(xué)工具解決 數(shù)據(jù)發(fā)掘 問題，并且促進(jìn)行業(yè)增長呢？我們從以下幾個(gè)主要行業(yè)進(jìn)行分析。

保險(xiǎn)業(yè)

以前的保險(xiǎn)公司是依靠人工進(jìn)行數(shù)據(jù)采樣的，除此以外，他們還要分析客戶群體并處理運(yùn)行中出現(xiàn)的各種問題。這一過程不僅費(fèi)時(shí)費(fèi)錢，也容易出現(xiàn)錯(cuò)誤。

人工分析依靠的是歷史數(shù)據(jù)，不可能對實(shí)時(shí)情況做出反應(yīng)，這意味著人工分析無法避免類似于詐騙這樣的威脅，因?yàn)檫@類威脅都是事情發(fā)生后才能發(fā)現(xiàn)問題。

據(jù)英國保險(xiǎn)業(yè)協(xié)會(huì)估計(jì)，每年未被發(fā)現(xiàn)的詐騙金額高達(dá)19億英鎊（約22億歐元，合163億人民幣），由此導(dǎo)致投保人每人每年要多花費(fèi)50英鎊保費(fèi)。根據(jù)2015年保險(xiǎn)詐騙調(diào)查報(bào)告顯示，詐騙集中在意外保險(xiǎn)（31%）和保險(xiǎn)申請（12%）這兩個(gè)方面。

現(xiàn)在保險(xiǎn)公司可以運(yùn)用更加先進(jìn)的分析學(xué)工具來避免這些威脅。通過分析學(xué)工具，英國的某汽車保險(xiǎn)公司幫助每名投保人節(jié)約了2英鎊的資金。瑞士的某保險(xiǎn)公司把不合規(guī)的相關(guān)風(fēng)險(xiǎn)降低到1%以下。

保險(xiǎn)業(yè)運(yùn)用分析學(xué)工具的好處：

• 風(fēng)險(xiǎn)定價(jià)更準(zhǔn)確；
• 可以更好地比對花費(fèi)與定價(jià)，來發(fā)現(xiàn)和維系客戶；
• 減少索賠中的失誤；
• 節(jié)約索賠決策時(shí)間；
• 通過分析不同的網(wǎng)絡(luò)平臺(tái)數(shù)據(jù)以減少詐騙的發(fā)生；
• 減少由失誤、不良債權(quán)、訴訟和客戶流失導(dǎo)致的成本費(fèi)用，以獲得更高利潤，提高客戶滿意度。

銀行業(yè)

金融機(jī)構(gòu)使用 數(shù)據(jù)分析 工具的動(dòng)機(jī)其實(shí)各不相同，互聯(lián)網(wǎng)數(shù)據(jù)中心（IDC）公布的一份調(diào)查結(jié)果向我們展示了這其中的區(qū)別。

?

但是不論他們的動(dòng)機(jī)如何，對銀行業(yè)來說分析學(xué)工具的優(yōu)勢是顯而易見的。 “美國銀行家研究”機(jī)構(gòu)（American Banker Research）對170名銀行家進(jìn)行了調(diào)查。其中28%的人認(rèn)為顧客份額是機(jī)構(gòu)獲得的最大收益，18%的人則認(rèn)為貸款虧損的減少是最關(guān)鍵的收益。

譯者注：顧客份額是一個(gè)企業(yè)為某一顧客所提供的產(chǎn)品和服務(wù)在該顧客同類產(chǎn)品和服務(wù)消費(fèi)總支出中所占的百分比。

很多銀行表示，最大程度地利用網(wǎng)上資源有益于維持客戶忠誠度。下圖展示了一家歐洲銀行依靠分析學(xué)工具，使得廣告點(diǎn)擊率提高了27%，銷售額提高了12%。

?

醫(yī)療業(yè)

近年來醫(yī)療領(lǐng)域出現(xiàn)了很多新變化，列舉其中幾個(gè)：

• 醫(yī)療業(yè)越來越趨向于價(jià)值導(dǎo)向，因?yàn)榭蛻魧Ω哔|(zhì)量的服務(wù)要求越來越高；
• 外科醫(yī)生和護(hù)士數(shù)量不足，這要求醫(yī)院設(shè)法提高員工的工作效率；
• 成本發(fā)生變化，因?yàn)樗劳雎式档土?，慢性病例?shù)量增多；
• 醫(yī)學(xué)研究獲得了更多的投資，因此新的藥物和療法不斷被發(fā)現(xiàn)。

這些變化增加了醫(yī)保提供者面臨情況的復(fù)雜性。運(yùn)用客戶導(dǎo)向分析工具，醫(yī)保提供者能夠輕易地實(shí)現(xiàn)以下功能：

• 發(fā)現(xiàn)疾病類型，防止疾病爆發(fā)，并快速對醫(yī)療緊急事故做出反應(yīng)；
• 跟蹤預(yù)防性治療，比如可以看到數(shù)據(jù)庫中有多少人接種過流感疫苗；
• 高效地分配有限的員工資源；
• 通過客戶在醫(yī)院網(wǎng)站的瀏覽歷史，預(yù)測他們的需求和病情；
• 減少浪費(fèi)，麥肯錫咨詢公司（McKinsey）預(yù)計(jì)，每年在臨床治療、研發(fā)和公共健康上，美國醫(yī)療行業(yè)能夠節(jié)省3000多億美元。

教育行業(yè)

教育行業(yè)擁有很多數(shù)據(jù)來源，比如：

• 錄取記錄（通常包含了社會(huì)經(jīng)濟(jì)數(shù)據(jù)、人口數(shù)據(jù)、歷史表現(xiàn)和健康情況數(shù)據(jù)等）；
• 實(shí)驗(yàn)室情況，圖書館、咖啡廳和一般消費(fèi)記錄；
• 出勤率、考試分?jǐn)?shù)和評分等級(jí)情況；
• 體育運(yùn)動(dòng)記錄。

但是，這些數(shù)據(jù)很難用于提高教育環(huán)境和預(yù)測學(xué)生需求。

亞利桑那州立大學(xué)率先運(yùn)用分析學(xué)，提升了用戶體驗(yàn)。該網(wǎng)站國際頁面收集的數(shù)據(jù)顯示，網(wǎng)站訪問量來自世界各地，這促使學(xué)校在網(wǎng)頁上提供了不同的語言選項(xiàng)。

電子商務(wù)

幾百萬個(gè)網(wǎng)站都爭先恐后地向同樣的客戶群體推銷商品，在這種情況下，如果不使用數(shù)據(jù)分析工具，商家就很難把握面臨的銷售環(huán)境和客戶情況。

電子商務(wù)企業(yè)有很多標(biāo)準(zhǔn)來衡量網(wǎng)絡(luò)性能，其中“轉(zhuǎn)換能力”是衡量性能的最關(guān)鍵的指標(biāo)（KPI）。對于那些轉(zhuǎn)換能力不高的網(wǎng)站可以進(jìn)行深度挖掘，發(fā)現(xiàn)背后的原因。很多優(yōu)秀的網(wǎng)站分析工具都可以做到這一點(diǎn)，這里列舉了一部分網(wǎng)站分析工具：

?

Wappalyzer整理出來的這份表格顯示，大部分網(wǎng)站都使用網(wǎng)站分析工具，如woocommerce.com 使用KISSmetrics，shutterstock.com使用crazy egg，app.hubstaff.com使用 woopra。

政府

盡管政府在產(chǎn)能和信息通訊技術(shù)上占主導(dǎo)地位，他們也在盡職盡責(zé)地投資開發(fā)報(bào)表工具、計(jì)算機(jī)設(shè)備和數(shù)據(jù)庫，但是在數(shù)據(jù)收集和定性分析上政府依然面臨著困難。

我們不僅要運(yùn)用分析學(xué)工具發(fā)掘數(shù)據(jù)，也要使用分析學(xué)工具提高分析質(zhì)量以解決問題。我們從數(shù)據(jù)中發(fā)掘的價(jià)值越多，就越能利用數(shù)據(jù)提高市民的生活質(zhì)量。

以美國政府網(wǎng)站為例，政府部門能夠看到一定時(shí)間內(nèi)，訪問其網(wǎng)站的人數(shù)、訪問的內(nèi)容以及下載的文件。這些數(shù)據(jù)清楚地顯示了人們所需要的政府服務(wù)。

?

結(jié)論

文中列舉的事例，展示了分析學(xué)工具的益處。這說明任何產(chǎn)業(yè)的發(fā)展契機(jī)，都依賴于其數(shù)據(jù)分析的能力。市場上已經(jīng)出現(xiàn)了很多性價(jià)比很高的分析工具，操作上也很簡便。這就意味著企業(yè)不需要復(fù)雜的數(shù)據(jù)收集和儲(chǔ)存基礎(chǔ)設(shè)施，就能夠輕松地使用他們的數(shù)據(jù)。

英文原文

How Analytics tools are shaping the growth story across industries

If there’s one thing that businesses across all industries have in common today, it’s in their increased adoption of data to shape business decisions. Below is a demonstration of how key industries use analytics tools and the benefits these tools have in solving challenges of data capture and use to shape growth.

Insurance

Traditionally, insurers have relied on manual sampling of data to understand their customer base and address challenges to their operations. Not only is this process time-consuming and costly, it is also highly prone to errors.

Manual analysis also relies on historical data, making it impossible to respond to changes that are happening in real-time. This means that threats such as fraud cannot be prevented, as they are only detectable after the fact.

The association of British insurers estimates the amount of annual undetected fraud at roughly ￡1.9bn (€2.2bn), a loss that costs policy holders an approximate cost increase of ￡50 on their yearly premiums. Some of the highest instances of fraud, according to the 2015 insurance fraud survey, are noted in staged accidents (31%) and applications (12%).

Insurers can protect themselves against such threats using better analytical insights. Below are two examples from Insurance Nexus of insurers who have benefited from use of claims analytics:

Annual savings of up to ￡2 in auto claims by a Uk insurer

A Swiss insurance company reduced risks associated with non-compliance to less than 1%

Benefits of analytics in underwriting:

Accuracy in risk pricing.

Identifying and retaining customers through better comparisons of costs and pricing.

Reductions of errors in claims.

Reduced decision making time where claims are concerned.

Reduced cases of fraud, through analysis of different web-based platforms.

Reduced costs associated with errors, bad claims, litigation and customer attrition, leading to more attractive margins and better customer satisfaction

Banking

Financial institutions differ in their motivations for investing in data analysis as shown in the survey results below conducted by IDC .

But whatever their motivations, the benefits of analytics in banking are clear. American Banker Research surveyed 170 bankers on the usage of customer analytics in banking. 28% of them cited share of wallet as the biggest benefit experienced by their institutions. Another 18% cited reduction in loan-related losses as the key benefit.

Banks also recognize the importance of making optimum use of their online resources to retain customer loyalty. As shown in the image below, a European bank experienced 27% increase in click-through rates for their banners and a sales increase of 12%, by relying on their analytics tools.

Healthcare

There are a lot of dynamics surrounding the field of healthcare. To mention, but a few:

Healthcare is becoming more value-based as customers continue to demand quality services.

Physicians and nurses are always in short supply which means that hospitals have to figure out how to be efficient and productive with the staff they have on hand.

Cost dynamics are changing, thanks to reduced death rates and more reported cases of chronic diseases.

More investment in research has led to new medical approaches and cures.

All these issues create a lot of complexity for health care providers. With more utilization of customer-based insights for decision making, healthcare providers will find it easier to:

Detect disease patterns, prevent outbreaks and respond to medical emergencies with speed.

Track implementation of preventive remedies. For instance, they can track how many people in their data base have received flu vaccines.

Efficient allocation of limited hospital staff.

Use customer browsing history on hospital websites to anticipate individual needs or crises.

Reduce wastage. Estimates by McKinsey show that U.S healthcare can reduce waste and save more than $300 billion annually in clinical operations, R&D and public health.

Education

Learning institutions have many data sources such as:

Admission/enrolment records, (which are usually a combination of socio-economic data, demographics, past performance, health issues, etcetera)

Laboratory, library, cafeteria and general purchase records,

Attendance, test scores and grade tracking,

Sports records.

However, this information is hardly used for improving the learning environment and to anticipate student needs.

Arizona state University is a good example of use of analytics by a learning institution to improve user experience.

Insights gathered from the website’s international page showed that most of the traffic to the website came from all over the world, a factor that prompted the university to offer the pages in different languages.

Ecommerce

Standing out among the millions of websites that are competing to sell to the same audience is near impossible without the use of data to help a business understand its environment and audience.

Though ecommerce businesses use many metrics to measure website performance, ‘conversion’ is the key KPI used to show the rate of success. A website that is experiencing low conversions can dig deeper to understand the reasons behind this performance. There are some amazing web analytics tools out there that you could choose from, here’s a list I’ve compiled about Web Analytics Tools which can prove to be handy.

Lists compiled by wappalyzer show that most key websites use web analytics tools. for instance, woocommerce.com uses KISSmetrics, shutterstock.com uses crazy egg and app.hubstaff.com uses woopra.

Government

Governments hold a central role in the ramping up and use of ICT and though they embrace this role fully by investing in reporting tools, computer equipment and data warehouses, there’s still a challenge when it comes to moving from mere data collection and processing to qualitative data analysis.

The use of tools to not only mine data but to improve analysis helps to address these challenges as the more value is extracted from data, the more it can be used to better the lives of citizens.

For instance, in the example below from the US government site, it’s possible for government departments to see how many people visit the website over a period of time, which pages they visit and the documents they download. This gives a clear indication of the services that people need most.

Conclusion

The body of evidence that shows benefits in increased investment in data tools suggests that opportunities for growth in any sector lie in data insights. The market has readily available analysis tools which are budget friendly and don’t require intensive training to operate, meaning that companies don’t need to roll out sophisticated data capture and warehousing infrastructure to start making use of their data.

注：本文摘自數(shù)據(jù)觀入駐自媒體—燈塔大數(shù)據(jù)，轉(zhuǎn)載請注明來源，微信搜索“數(shù)據(jù)觀”獲取更多大數(shù)據(jù)資訊。

?

責(zé)任編輯：陳卓陽