《SoftServe&谷歌云：2025 AI時代構建主動數據治理框架白皮書：化數據負債為戰略資產（英文版）（33頁）.pdf》由會員分享，可在線閱讀，更多相關《SoftServe&谷歌云：2025 AI時代構建主動數據治理框架白皮書：化數據負債為戰略資產（英文版）（33頁）.pdf（33頁珍藏版）》請在三個皮匠報告上搜索。

1、TURN YOUR DATA FROM LIABILITY TO ASSETBuilding an Intelligence FrameworkIn todays rapidly evolving digital landscape,data is no longer just a byproduct of business operations;its a strategic asset that fuels innovation,customer satisfaction,and competitive advantage.As you explore this eBook,conside

2、r how SoftServes expertise and Googles innovations can help you enable durable data governance that supports turning data into an asset,safely.Together,we help organizations:Prioritize data quality to improve ROI Develop real-time analytics dashboards for actionable insights Democratize data to empo

3、wer decision-making Future-proof your data in an AI-enabled worldThe Intelligence Framework from SoftServe,coupled with Google Clouds advanced data and AI capabilities,helps businesses access the full potential of their data and accelerate AI initiatives to achieve tangible business value.Turn your

4、datafrom liability to assetHow to build proactive data governance in the AI eraContents03 Introduction04 Why data governance matters in the era of AI.Three ways to achieve more proactive data governance07 Prioritize data quality to drive ROI.Build trust in your data.Empower data-driven decisions.Imp

5、roveAIperformance.How industries are empowering data-driven decisions Next steps14 Turn non-technical users into data-driven decision makers.Unleash insights.Securely automate self-service.ImproveAIperformance.How industries are empowering data-driven decisions Next steps20 Rise above architecture.E

6、asily adapt to regulatory updates.Achieve compliance across architectures.Unify your data with a metadata lens.How automation delivers results Next steps27 Banish dark,duplicate,and dirty data to maximize ROI.32 EndnotesYour guide to a data governance strategy to maximize the value of your dataData

7、can be a liability.It costs to store it.It costs to safeguard it.And with the average financial toll of a data breach estimated at$4.35 million,1 it costs even more to lose control of it.But data is also your organizations strongest asset no matter the size of your organization or which industry you

8、 are in.Data is the key to discovering insights,informing great customer experiences,and driving innovation.Data governance is a principled approach to managing data during its life cycle from acquisition to use to disposal.Adurable data governance framework unlocks your data asset.From processes th

9、at unify your data across disparate architectures to policies that enable employees to access information when needed,data governance enables your organization to extract long-term ROI from your data and implement AI transformation.To maximize your data asset,your organization requires proactive and

10、 automated data governance with built-indata intelligence.Introduction3|Turn your data from liability to asset|IntroductionOrganizations have more data than ever before.And keeping it secure and compliant while extracting valuable insights is an increasingly complex task.Data governance is a set of

11、processes and policies that ensure effective data management within an enterprise organization.And automated,AI-driven data governance is the key to tackling todays data quality,democratization,and compliance challenges at scale.Organizations that dont take advantage will soon be left behind.Learn m

12、ore about data governanceWhy data governancemattersin the era of AIData&ML OpsAuto-discover data and metadataTrustwith lineage and qualityCurateper organization policiesOrganizedata into data domainsMonitorand audit changesEnrichbusiness contextSecurewith active metadata-driven policies4|Turn your d

13、ata from liability to asset|Why data governance matters in the era of AIBy 2030,companies that fully absorb AI could double their cash flow,while companies that dont could see a20%decline.25|Turn your data from liability to asset|Why data governance matters in the era of AIOrganizations need to demo

14、cratize data while enabling domain-specific and corporate-level governance.Tools with built-in data intelligence enable organizations to achieve end-to-end visibility and management of their data.From there,automated tools and policies can grant access to relevant insights to employees across the or

15、ganization while ensuring compliance with rapidly changing regulations.While data governance is commonly perceived as safeguarding data,its also key to helping organizations build whats next.Quality data drives business intelligence and informs technologies such as generative AI and large language m

16、odels(LLMs).Aproactive data governance practice puts data at the core of your business and powers innovation.This guide shows you how to build a future-proof data governance strategy that supports data throughout its life cycle,using automated principles,practices,and technologies.Well show you thre

17、e ways to get started.Manage your data at scale.With a proactive data governance strategy,your organization can:Ensure data quality.Make trusted data easily discoverable.Provide insights into data origination and lineage.Securely govern data sharing internally and externally.Meet regulatory requirem

18、ents.6|Turn your data from liability to asset|Why data governance matters in the era of AIThree ways to achieve more proactive data governance1 Prioritize data quality to drive ROI.7|Turn your data from liability to asset|Prioritize data quality to drive ROIData is valuable if you use it well.Achiev

19、ing ROI from your organizations data relies on trusted,quality data enabling you to drive real-time decisions,power effective AI models,and discover opportunities for better customer experiences.So you can:Build trust in your data.Empower data-driven decisions.Improve AI performance.By 2025,Global d

20、ata creation is projected to reach more than 180 zettabytes.3(1 zettabyte=1 trillion gigabytes)8|Turn your data from liability to asset|Prioritize data quality to drive ROIBuild trust in your data.With rising volumes of data spread across silos,it can be challenging for organizations to achieve an e

21、nd-to-end view of their data.But when organizations comprehensively understand data quality,they can trust their data.So that when users find the data theyre looking for,theyre assured that its up-to-date,accurate,and complete.Using automated monitoring of how data moves through your systems(data li

22、neage),data observability tools can detect and repair data anomalies.If theres a spike or drop in the data,these tools help you determine whether the data is complete or missing a record.Andif there is a data issue,producers or data engineers can be alerted as it happens.Empower data-driven decision

23、s.Data quality is the foundation of the business intelligence(BI)that enables your organization to deliver high-value user experiences.Business intelligence can uncover critical patterns and trends in nearly every area of your organization including sales,customer service,production,security,and mor

24、e.Real-time analysis enables organizations to discover actionable insights that improve short and long-term performance,driving ROI organization-wide.The more data you collect and analyze,the more security and governance matter.9|Turn your data from liability to asset|Prioritize data quality to driv

25、e ROIData powers real-time targeted recommendations in retail to deliver a personalized online shopping experience.And provides faster access to applications to reduce latency in online transactions.10|Turn your data from liability to asset|Prioritize data quality to drive ROIImprove AI performance.

26、Trust(or lack thereof)in AI development is a leading barrier to adoption.While there are numerous considerations for building more trusted AI models,data quality is foundational.For example,if training data contains unfair biases,these biases will also be reflected in the model.In cases where AI sys

27、tems make decisions that significantly impact individuals such as loan approvals,student grades,or employee hiring decisions organizations that curate their training data must be accountable for these AI systems decisions and prevent illegal discrimination.Data quality is a prerequisite to creating

28、responsible AI models.And data governance practices are how organizations ensure their data adheres to quality standards and can be trusted.The reliability of your data is the groundwork for AI innovation.While complete considerations for AI models is outside the scope of this paper,here are some th

29、ought starters:Generative AI foundation models come pre-trained on large corpora of data.When deploying gen AI,you trust that the foundation model is unbiased.Because of this,its wise to find a provider that offers you indemnity for any claims related to the training data of the model you use.Qualit

30、y AI models rely on more than just training data.Its important to groundyourmodels with data at the core.Embeddings make use of semantic relationships between your data,so that your model delivers accurate answers,not just the probabilistic answers that are typical output of an LLM.Human intelligenc

31、e plays a key role in monitoring AI systems for bias while a comprehensive data governance approach ensures trustworthy data is the foundation for AI training.11|Turn your data from liability to asset|Prioritize data quality to drive ROIFinancial servicesMacquarie Bank brings hyper-personalized serv

32、ices to its retail banking customers,including auto categorization that gives customers a holistic view of their transactions and expenses so they can manage their finances more effectively.HealthcareJohns Hopkins University BIOS Division has achieved faster and more accurate decision-making for bra

33、in injury patients,lowering the time to evaluate 500 scans from 2,500 hours to 90 minutes and reducing research and infrastructure costs by 50%.How industries are empowering data-driven decisions12|Turn your data from liability to asset|Prioritize data quality to drive ROIWhen assessing the quality

34、of their data,many organizations initially struggle with questions such as:Is the data extracted from an authoritative source?When was this data last refreshed?How is sensitive information being moved or copied?Is it in adherence to data governance practices?Organizations can improve data quality th

35、rough data observability by monitoring their data and proactively addressing any issues:Implement data lineage tools to track how data is sourced and transformed,including sensitive data.Put granular and broader checks in place for an ecosystem-wide view,delivering the right alerts and reports.When

36、data is flagged,quarantine that data and take action to fix it before its a problem.These proactive practices ensure that when your users access data,it can be trusted.This level of trust is foundational for BI insights and data democratization that drive ROI.Next steps13|Turn your data from liabili

37、ty to asset|Prioritize data quality to drive ROIThree ways to achieve more proactive data governance2 Turn non-technical users into data-driven decision makers.Data teams are a common bottleneck.Data democratization giving people access to insights and the know-how they need to work with data profic

38、iently can relieve the pressure.However,many organizations need help maintaining consistent governance while enabling self-service of data.They need a framework that provides high risk and cost mitigation while providing broad access to data that teams need.Secure self-access tools and processes bac

39、ked by strong data governance policies are the keys to resolving the tension between access and security.So you can:Unleash insights.Securely automate self-service to data.Increase innovation.69%of data professionals spend an average of 610 hours per week responding to,managing,and resolving data ac

40、cess issues.Thats 2440 hours a month,or 288480 hours a year.4 15|Turn your data from liability to asset|Turn non-technical users into data-driven decision makersUnleash insights.Business intelligence doesnt come from simply making data accessible in big and unstructured data.The democratization of i

41、nsights(the meaning and implications of the data)leads tovalue generation.Imagine a renewals manager reviewing a customers profile in your CRM before an important call.With access to data about that customer,the employee can glance through data tables to see information about how the customer uses y

42、our software platform and try to glean something useful.Bycontrast,with access to insights about that account,the renewals manager could instantly see that the customer has continually hit their usage limit and is a likely upsell candidate for more seats.AI-powered,embedded analytics tools give peop

43、le access to the insights they need when they need them.And with the right integrations,employees can access these analytics in the workflows theyre accustomed to even outside your BI platform.Analytics shouldnt be limited to technical users or data engineers.Todays low-code tools empower employees

44、who need more technical skills to access insights.In the near future,natural language processing and gen AI will be the gold standard for real-time search capabilities.Learn moreabout building a data-driven culture16|Turn your data from liability to asset|Turn non-technical users into data-driven de

45、cision makersSecurely automate self-service.Data is critical to understanding your organizations customers,market,and competitors with self-service unlocking increased potential for insights.However,without the right security controls,self-service can present increased risk.Automated processes can k

46、eep your customers data secure and compliant,while granting autonomy to the user closest to certain data types.Domain-based architectures empower individual business teams to own a specific use case or data set.A strong data governance framework gives these teams access to and governance of that dat

47、a while keeping the umbrella set of compliance controls in place.Increase innovation.Modern organizations are data-driven.Not just to enable better customer experiences and outcomes,but to reimagine their organizations for what comes next.When it comes to innovating in a rapidly changing world,colla

48、boration is a key factor that sets leading organizations apart.6Data democratization breaks down organizational silos and enables a larger pool of employees to use data to solve problems and develop new ideas.And with lower barriers to collaborating,sharing,and benefiting from each others work,emplo

49、yees can collaboratively drive a bigger impact.96%of digital leaders had C-suite executives that empowered teams to generate ideas and own decisions,while 73%of laggards had leadership predicated on following instructions.569%of employees bypassed their organizations cybersecurity guidance in the pa

50、st 12 months.74%of employees said they would bypass cybersecurity guidance if it helped them or their team achieve a business objective.617|Turn your data from liability to asset|Turn non-technical users into data-driven decision makersImagine shortening the development time of custom reports from w

51、eeks to days,and freeing up significant engineering resources in the process.Thats the power of democratization,and while its the future for many companies,Subskribe is already there.Subskribe provides a unified quote-to-revenue platform for modern SaaS platforms,helping them improve decision-making

52、 and drive growth with on-demand insights.Previously,onboarding a new SaaS customer required weeks of engineering effort to develop custom reports.By integrating Looker Embedded,non-technical teams can now develop sophisticated reports without code,freeing engineering up for innovation opportunities

53、.But thats just the beginning.Subskribe is now building an embedded analytics solution with Looker that will enable customers to create their own dashboards and reports.And expanded personalization options will serve up trusted analytics that are tailored for user roles such as executives,finance st

54、aff,and client success teams driving product adoption andcustomer success.How data democratization ischangingthegameWe designed our semantic model so that,in just a matter of hours,anyone can use Looker to build their own dashboards and reports with the data theyreauthorized to see.”Ugurcan Aktepe S

55、oftware Engineer,Subskribe18|Turn your data from liability to asset|Turn non-technical users into data-driven decision makersCollaboration is key to spurring innovation.And traditional barriers to collaboration dispersed teams and siloed information are dissolving in a connected and democratized mod

56、ern workplace.However,security remains an essential concern.Employees shouldnt have to worry about who has access to what information when collaborating.Organizations can fuel collaboration by using automated processes and tools to ensure its secure,by design.97%of industry leaders say democratizing

57、 access to data and analytics across the organization is important to business success.7 A data democratization framework with security built-in:Protects data at every step(with policies for how data is accessed,managed,and retained).Maintains visibility into each data action(using data logs and lin

58、eage tools).Makes it easy for users to do the right thing(with automatic protections and tools that help employees share information securely).Embedded analytics in Looker meets users where they are in skills,requirements,and preferences.Learn more in this Google Cloud webinarNext stepsData democrat

59、ization means building an organization-wide culture and processes that sustainably supports self-service access to data that is compliant and secure.”Prajakta Damle Director of Product Management,Google Cloud19|Turn your data from liability to asset|Turn non-technical users into data-driven decision

60、 makersThree ways to achieve more proactive data governance3 Rise above architecture.20|Turn your data from liability to asset|Riseabove architectureAs data regulations increase globally,organizations are also grappling to protect more complex architectures and data access needs.With many enterprise

61、s working in hybrid across on-premises and cloud or multiple clouds,its clear that compliance can no longer be restricted to individual architectures.Compliance requires a proactive,end-to-end strategy with umbrella policies indifferent to infrastructure.So you can:Easily adapt to regulatory updates

62、.Achieve compliance across architectures.Unify your data with a metadata lens.Countries are introducing data regulations rapidly,leaving companies struggling to remain compliant.8 21|Turn your data from liability to asset|Riseabove architectureEasily adapt to regulatory updates.Data regulations are

63、increasing in scope across geographies and industries.Once the realm of data-intensive industries,today even small companies are bound to government privacy and security regulations that keep customer data secure.When compliance is an add-on(reactive),each new regulation complicates your data govern

64、ance framework.But when compliance is built-in(proactive)and based on business-driven processes,your organization remains end-to-end compliant and can manage regulatory updates with ease.Achieve compliance across architectures.Many organizations operate across multiple clouds and architectures to pr

65、ovide critical services and protect sensitive information.Extensibility enables your organization to remain secure and compliant.And open source enables integration with partners that can provide extensibility,end-to-end.Open source improves extensibility.Proprietary software is plagued with interop

66、erability issues and presents survivability challenges.Solutions based on open source tools and open standards are easier to integrate,alleviate a significant portion of development cost,and give organizations the flexibility to deploy and,if necessary,migrate critical workloads across or even off p

67、ublic cloud platforms.Having tools that connect across architectures is key to gathering functional insights while remaining compliant.The Cloud Data Management Capabilities(CDMC)framework defines the key controls required to protect data in cloud environments and meet data privacy requirements arou

68、nd the world.Google Clouds Digital Sovereignty Explorer is an online tool that takes individuals through a series of questions about their organizations digital sovereignty requirements.99%of companies are using open source and 35%of all enterprise software is based on open source code.11 2022 saw t

69、he most downloads of open source data ever.12 22|Turn your data from liability to asset|Riseabove architectureUnify your data with a metadata lens.Data is key.But first,your organization needs to know where it is.AI-driven metadata tagging enables organizations to discover and harness structured and

70、 unstructured data,while improving their understanding of data lineage.Traditionally,analytics were optimized for data warehouses structured architectures that required the data to be extracted,transformed,and loaded into a specific format.Organizations can now take advantage of less rigid architect

71、ures,but without end-to-end discoverability,dark data is a looming liability.Automation is key to properly tag,store,and dispose of data to adhere to data sovereignty regulations and data democratization compliance.Implementing your metadata tagging and annotation strategy requires more than just te

72、chnology.Automation can discover well-known data sets,but the specific business context is still something the organization needs to know and define,and make part of its data governance strategy.Data that is collected but not used is called dark data.This term includes undiscovered data,underutilize

73、d data,and private identifying information(PII)that is improperly categorized.of organizations reported that at least half of their data is dark,9 posing significant risk.66%23|Turn your data from liability to asset|Riseabove architectureGoogle Cloud supports retention policies across a portfolio.So

74、 that when the time comes,Google Cloud will move data or delete it including the log that indicates what has happened.24|Turn your data from liability to asset|Riseabove architectureDeutsche Bank unlocked a key driver for accelerating transformation by automating controls.Achievers,with over 3.6 mil

75、lion users across 190 countries,have maintained their data quality while rapidly growing to provide data trust across the organization,saving hundreds of hours across the team.Casa Dos Ventos went from frequent“data firefighting”to automatic data quality checks that build trust in their data,provide

76、 the context needed to set the right efficiency goals,and help them invest wisely in the Brazilian renewable energy market.How automation delivers results25|Turn your data from liability to asset|Riseabove architectureData governance informs the management of the entire life cycle of your data,inclu

77、ding its disposal.Many jurisdictions require that personal information be destroyed after a set amount of time.However,the patchwork of regulations and the fragmented nature of organizations data can make compliance technically challenging.2Next stepsDiscover your data.With more data than ever befor

78、e,its important to not only know where data is located,but identify whether it is sensitive information that is subject to stricter security and compliance policies.This visibility requires end-to-end discoverability and lineage tools.Annotate.Classify your data using your business context to create

79、 meaningful business domains.Your data governance strategy can then follow from the lens of how you see your business and not how you see the infrastructure.Apply automated governance policies.Express your governance intent based on your annotations and apply those policies at scale across distribut

80、ed data to secure and govern your data and ensure compliance to different regulations.Defining these policies based on your business context and automating these policies ensures your data is accessed securely,purged when required,and also retains audit logging to report to regulators.Here are 3 ste

81、ps to ensure your organization remains compliant:26|Turn your data from liability to asset|Riseabove architectureBanish dark,duplicate,and dirty data to maximize ROI.27|Turn your data from liability to asset|Banish dark,duplicate,and dirty data to maximize ROIYour organization is collecting data at

82、levels like never before.You have the potential to leverage this asset to become a data-driven organization that delivers exceptional customer experiences and innovates for whats next.But too often,data is a liability.Dark data.Duplicate data.Dirty data.All can cause havoc and prevent your organizat

83、ion from realizing your datas potential.A comprehensive data governance strategy removes the roadblocks to being data-driven.By ensuring data quality,data security,and data compliance,your organization can unlock its potential.28|Turn your data from liability to asset|Banish dark,duplicate,and dirty

84、 data to maximize ROI“Unstructured data from chat applications or log files can cause significant headaches for organizations,especially if they unexpectedly contain sensitive data like PII.An example of this is customer support transcripts,because you never know what information people will submit.

85、When someone chats with customer support,they could type,I didnt get my medications.Heres my name,the medications Ineed,and my social security number.That sensitive PII data isnow in one of your databases which may not be appropriately secured and classified.”Anton Chuvakin,Senior Staff Security Con

86、sultant,Google Cloud(Data&AI Trends)29|Turn your data from liability to asset|Banish dark,duplicate,and dirty data to maximize ROIReady to turn your data from liability to asset?Weve covered how organizations make the most oftheir data in the AI era,including how to:Break down data silos.Improve AI

87、performance.Build trust in their data.Democratize access to insights.Improve collaboration and innovation.Ensure compliance across architectures.Modern organizations are embracing proactive,AI-driven,built-in approaches to data governance where employees can access the insights they need,customer da

88、ta is kept secure,and the organization is set to embrace the AI transformation.The data is already yours.Discover it.Democratize it.Harness it.30|Turn your data from liability to asset|Banish dark,duplicate,and dirty data to maximize ROI1 Cost of a data breach report(2023)IBM.2 The consumer-data opp

89、ortunity and the privacy imperative(2020)McKinsey.4 2023 State of Data Engineering Survey(2023)Immuta.5 Gartner Predicts Nearly Half of Cybersecurity Leaders Will Change Jobs by 2025(2023)Gartner.6 The Keys to Scaling Digital Value(2022)Boston Consulting Group,sponsored by Google.7 Turning Data into

90、 Unmatched Business Value(2023)Harvard Business Review,sponsored by Google.8 This is what increasing data protection laws mean for your company(2023)World Economic Forum.9 Dont Let Data(In)Visibility Limit Your Digitization Dreams(2022)Gartner.10 The consumer-data opportunity and the privacy imperat

91、ive(2020)McKinsey.11 How leading organizations are making open source their super power(2022)Google Next 22.12 State of the Software Supply Chain(2023)Sonatype.Endnotes LEARN MORESoftServe offers:Deep expertise:SoftServes hundreds of Google-certified experts have extensive experience in data strateg

92、y,AI/ML,and Google Cloud technologies.Tailored approach:ROI-prioritized and customized solutions address specific business needs and data challenges to ensure each organization achieves the desired business outcomes.Proven results:SoftServe has an extensive track record of successful data and AI imp

93、lementations with Google Cloud.Learn more today about how we can help you maximize your data assets.SoftServe helps organizations to rapidly deploy,utilize,and scale Google Cloud products with greater speed and security.Prioritized,tailored solutions that address industry-specific use cases get you to business value faster.